Skip to main content

Full text of "Journal of the Society of Motion Picture Engineers"

See other formats





JANUARY, 1934 


The Society of Motion Picture Engineers 

Its Aims and Accomplishments 

The Society was founded in 1916, its purpose as expressed in its 
constitution being the "advancement in the theory and practice of mo- 
tion picture engineering and the allied arts and sciences, the standardi- 
zation of the mechanisms and practices employed therein, and the 
maintenance of a high professional standing among its members." 

The membership of the Society is composed of the best technical 
experts in the various research laboratories and other engineering 
branches of the industry in the country, as well as executives in the 
manufacturing and producing branches. 

The Society holds two conventions a year, spring and fall, at various 
places and generally lasting four days. At these meetings papers 
dealing with all phases of the industry theoretical, technical, and 
practical are presented and discussed and equipment and methods 
are often demonstrated. A wide range of subjects is covered, many 
of the authors being the highest authorities in their particular lines of 

Papers presented at conventions, together with contributed arti- 
cles, translations and reprints, abstracts and abridgments, and other 
material of interest to the motion picture engineer are published 
monthly in the JOURNAL of the Society. The publications of the 
Society constitute the most complete existing technical library of 
the motion picture industry. 




Volume XXII JANUARY, 1934 Number 1 



Report of the Committee on Laboratory and Exchange Practice 3 

Report of the Projection Practice Committee 11 

Report of the Historical and Museum Committee 13 

Report of the Committee on Standards and Nomenclature 17 

Sprocket Dimensions for 35-Mm. Visual and Sound Projection 

Equipment H. GRIFFIN 20 

Direct-Current High-Intensity Arcs with Non-Rotating Positive 

Carbons D. B. JOY AND A. C. DOWNES 42 

A New Development in Carbon Arc Lighting P. MOLE 51 

A New White Flame Carbon for Photographic Light 


The Use of the Talking Picture as an Additional Educational 

Tool at the University of Chicago H. B. LEMON 62 

A New 35-Mm. Portable Sound Projector H. GRIFFIN 70 

Society Announcements 78 





Board of Editors 
J. I. CRABTREE, Chairman 


Subscription to non-members, $8.00 per annum; to members, $5.00 per annum, 
included in their annual membership dues; single copies, $1.00. A discount 
on subscriptions or single copies of 15 per cent is allowed to accredited agencies. 
Order from the Society of Motion Picture Engineers, Inc., 20th and Northampton 
Sts., Easton, Pa., or 33 W. 42nd St., New York, N. Y. 

Published monthly at Easton, Pa., by the Society of Motion Picture Engineers. 

Publication Office, 20th & Northampton Sts., Easton, Pa. 
General and Editorial Office, 33 West 42nd St., New York, N. Y. 
Entered as second class matter January 15, 1930, at the Post Office at Easton, 
Pa., under the Act of March 3, 1879. Copyrighted, 1934, by the Society of 
Motion Picture Engineers, Inc. 

Papers appearing in this Journal may be reprinted, abstracted, or abridged 
provided credit is given to the Journal of the Society of Motion Picture Engineers 
and to the author, or authors, of the papers in question. The Society is not re- 
sponsible for statements made by authors. 

Officers of the Society 

President: A. N. GOLDSMITH, 444 Madison Ave., New York, N. Y. 
Past President: J. I. CRABTREE, Kodak Park, Rochester, N. Y. 
Vice-President: W. C. KUNZMANN, Box 400, Cleveland, Ohio. 
Vice-President: O. M. GLUNT, 463 West St., New York, N. Y. 
Secretary: J. H. KURLANDER, 2 Clearfield Ave., Bloomfield, N. J. 
Treasurer: T. E. SHEA, 463 West St., New York, N. Y. 


E. COUR, 1029 S. Wabash Ave., Chicago, 111. 
H. T. COWLING, 7510 N. Ashland Ave., Chicago, 111. 
R. E. FARNHAM, Nela Park, Cleveland, Ohio. 
H. GRIFFIN, 90 Gold St., New York, N. Y. 
E. HUSE, 6706 Santa Monica Blvd., Hollywood, Calif. 
W. B. RAYTON, Rochester, N. Y. 
H. G. TASKER, 41-39 38th St., Long Island City, N. Y. 


The first report of this Committee discussed laboratory procedure 
in general. The discussion is continued in this report, but now is 
restricted to the development of the film, which is divided into three 
phases, viz., (1) the preparation of a negative developing solution, 
(2) the preparation of a positive developing solution, and (3) means 
of development control. The material has been selected from the 
accompanying bibliography supplemented by information collected 
by the members of the Committee. 


The D-76 borax developer for motion picture negative film was 
first used for developing duplicate negatives. The excessive graini- 
ness of duplicate prints led to the substitution of this fine-grain 
developer for the old sodium carbonate developer. As the results 
attained with the new developer were very satisfactory, it was soon 
tried also for developing regular negatives. Several papers were 
published describing the characteristics of the borax developer, and 
the film manufacturing companies began to recommend its use by 
the motion picture laboratories. Today it is used, variously modified, 
in practically every motion picture laboratory. 

The original borax formula was prepared for rack-and-tank de- 
velopment. In most instances it proved to be too strong a solution 
to be used in developing machines, where the high speed of the film 
and the forced circulation of the developer caused a decided increase 
in the degree of agitation. 

Due to the lack of standardization, very little uniformity obtains 
in the construction and operation of the developing machines in- 
stalled in the various laboratories. Each machine must be operated 
in a particular specified manner if it is to be operated efficiently and 
the cost of developing a unit length of film be maintained at a mini- 
mum. This necessitates the most careful use of the developer to 
limit the cost of the chemicals, and the operation of the developing 

* Presented at the Fall, 1933, Meeting at Chicago, 111. 


machine at the maximum safe speed to limit the cost of labor and 
equipment. Therefore, it has been necessary to prepare a different 
formula in almost every laboratory, in order that the desired contrast 
and density might be obtained under each peculiar set of operating 
conditions. It may be added that these desired results are not yet 
completely standardized, as good photographic quality is still a sub- 
ject of personal judgment. 

With the assistance of the photographic chemists, the laboratory 
technicians have been readily able to modify the standard borax 
formula, and prepare solutions suitable for the special operating 
conditions. Two methods are available for finding the proper 
formula: the standard solution can be diluted, or the relative quan- 
tities of the constituents can be varied. Considerable danger is 
incurred if dilution is carried too far. Negatives of poor quality 
may result from a change in the characteristic curve of the developer ; 
or the efficiency of the solution may be decreased, with an accom- 
panying increase in the cost of chemicals. 

For these reasons a developer of the proper characteristics is usu- 
ally found by varying the constituents of the standard developer. 
A solution of the standard formula is first put into the machine, and 
the contrast and density are checked with the machine operating 
under the desired conditions. Additional chemicals in solution are then 
added until the desired contrast and density are attained. If it is 
necessary to vary the standard formula to such an extent that the 
efficiency of the developer has been appreciably decreased, it may 
be necessary to vary the operation of the machine or to alter its 
design. Such changes, of course, would be necessary only in the 
case of machines designed and constructed in the early days of ma- 
chine developing, when little was known of their operating speeds 
and capacity. 

This method of varying the constituents of the borax developer 
so as to achieve certain desired characteristics has been discussed 
in several papers presented to the Society. In particular, the paper 
entitled "Some Properties of Fine Grain Developers for Motion 
Picture Film," by Carlton and Crabtree,* has proved extremely 
helpful to the laboratory technicians in preparing satisfactory nega- 
tive developing solutions. It demonstrates in detail the numerous 
changes that occur in the characteristics of the solution when the 

* See appended bibliography, under Developing Solutions. 


quantities of the chemicals are changed, as well as the effect of 
adding other chemicals. 

In general practice the same formula is used for the negative 
replenisher as for the original solution. In some laboratories, how- 
ever, the negative replenisher is much more concentrated than 
the developing solution. The necessary rate of flow of additional 
solution to maintain the bath at constant strength is readily deter- 
mined by sensi tome trie tests. 


The D-16 developer for motion picture positive film was recom- 
mended to the laboratories many years prior to the era of machine 
development. It still serves today as a very satisfactory positive 
solution for preliminary testing in selecting a positive solution suit- 
able for a particular developing machine. Just as when preparing 
the negative developer, preliminary tests are made with the standard 
developer, and the quantities of the constituents are varied until 
the desired results are attained. Equal percentage variations of the 
metol or monomethylparaminophenol sulfate, hydroquinone, and 
sodium carbonate may be made over a wide range, with no appre- 
ciable decrease in the efficiency of the developer or change in its 
characteristic curve. In general, a change in the quantity of metol 
will cause a greater change in the density than the corresponding 
change in the contrast. Similarly, a change in the quantity of 
hydroquinone will cause a greater variation in the contrast than in 
the density. As has been mentioned, the density produced by the 
positive bath for a standard exposure must remain constant if 
prints are to be made from old negatives using the original timing 

The laboratories on the West Coast favor a special formula for the 
positive replenishing solution. In the eastern section of the country 
the formula of the replenisher is the same as that of the original 
solution, except that the potassium bromide is omitted. Both seem 
to produce satisfactory results. The problem of properly selecting 
a positive replenishing solution has been ably discussed by Crabtree 
and Ives in a paper entitled "A Replenishing Solution for a Motion 
Picture Positive Film Developer."* 

* See appended bibliography, under Developing Solutions. 



More progress has been made in development control during the 
past several years than in any other phase of laboratory work. To 
a great extent this is due to improvements that have been made in 
sensitometric equipment, and the consulting service on sensitometry 
now available to the laboratories. The Eastman type lib sensito- 
meter and the Eastman densitometer, which have been described in 
the JOURNAL,* have proved very satisfactory for laboratory control 

The methods used by the Hollywood laboratories in applying these 
instruments were described to the Society by Mr. E. Huse in his 
paper "Sensitometric Control in the Processing of Motion Picture 
Film in Hollywood."* The Committee has found that the same 
methods, with very little variation, are used in all other sections 
of the country also. 

R. F. NICHOLSON, Chairman 









Developing Machines and Associated Equipment 

"Erbograph Machine," R. C. Hubbard, Trans. Soc. Mot. Pict. Eng., VII 
(1923), No. 17, p. 163. 

"Straight Line Developing Machine," R. C. Hubbard, Trans. Soc. Mot. Pict. 
Eng., VIII (1924), No. 18, p. 73. 

"Machine Developing of Negative and Positive Motion Picture Film," A. B. 
Hitchins, Trans. Soc. Mot. Pict. Eng., IX (1925), No. 22, p. 46. 

"A Negative Developing Machine," C. R. Hunter, Trans. Soc. Mot. Pict. 
Eng., XII (1928), No. 33, p. 195. 

"A Horizontal Tray Type of Continuous Processing Machine," H. T. Jamieson, 
Trans. Soc. Mot. Pict. Eng., XII (1928), No. 36, p. 1093. 

"A Method of Quantity Developing of Motion Picture Film," C. R. Hunter 
and R. M. Pierce, /. Soc. Mot. Pict. Eng., XVII (Dec., 1931), No. 6, p. 954. 

"Rust Proof Steel Tanks for Developers," Lichtbild (June, 1932), No. 11, p, 

"Syphons and Measuring Devices for Photographic Solutions," K. C. D. 
Hickman, Trans. Soc. Mot. Pict. Eng., X (1926), No. 26, p. 37. 

* See appended bibliography, under Development Control. 


"Rack Marks and Airbell Markings on Motion Picture Film," J. I. Crabtree 
and C. E. Ives, Trans. Soc. Mot. Pict. Eng., IX (1925), No. 24, p. 95. 

"The Handling of Motion Picture Film under Various Climatic Conditions," 
R. J. Flaherty, Trans. Soc. Mot. Pict. Eng., X (1926), No. 26, p. 85. 

"The Examination of Film by Projection for a Continuous Processing 
Machine," W. V. D. Kelley, Trans. Soc. Mot. Pict. Eng., XI (1927), No. 30, 
p. 224. 

"Handling of Motion Picture Film at High Temperatures," J. I. Crabtree, 
Trans. Soc. Mot. Pict. Eng., VIII (1924), No. 19, p. 39. 

"Directional Effects in Continuous Film Processing," J. Crabtree^ /. Soc. 
Mot. Pict. Eng., XVIII (Feb., 1932), No. 2, p. 207; /. Soc. Mot. Pict. Eng., XXI 
(Nov., 1933), No. 5, p. 351. 

"A Modern Laboratory for the Study of Sound Picture Problems," T. E. 
Shea, /. Soc. Mot. Pict. Eng., XV (March, 1931), No. 3, p. 277. 

"Materials for the Construction of Motion Picture Processing Apparatus," 
J. I. Crabtree, G. E. Matthews, and J. F. Ross, /. Soc. Mot. Pict. Eng., XVI 
(March, 1931), No. 3, p. 330. 

"Movement in Film Tanks," Brit. J. Phot., LXXIX (July 8, 1932), No. 3766, 
p. 405. 

"Individual or Multiple Film Development," C. Emmerman, Photofreund, 
XII (March, 1932), No. 5, p. 86. 

Developing Solutions 

"A Preliminary Note on the Development of Motion Picture Film," F. F. 
Renwick, Trans. Soc. Mot. Pict. Eng., VII (1923), No. 16, p. 159. 

"The Development of Motion Picture Films by the Reel and Tank System," 
J. I. Crabtree, Trans. Soc. Mot. Pict. Eng., VII (1923), No. 16, p. 163. 

"Investigation on Photographic Developers," J. I. Crabtree and M. L. Dundon, 
Trans. Soc. Mot. Pict. Eng., VIII (1924), No. 19, p. 28. 

"Behavior of Gelatin in the Processing of Motion Picture Film," S. E. Shep- 
pard, Trans. Soc. Mot. Pict. Eng., XI (1927), No. 32, p. 707. 

"Motion Picture Photomicrographs of the Progress of Development of a 
Photographic Image," C. Tuttle and A. P. H. Trivelli, Trans. Soc. Mot. Pict. 
Eng., XII (1928), No. 33, p. 157. 

"The Fogging Properties of Developers," M. L. Dundon and J. I. Crabtree, 
Trans. Soc. Mot. Pict. Eng., XII (1928), No. 36, p. 1096. 

"Some Properties of Fine Grain Developers for Motion Picture Film," H. C. 
Carlton and J. L Crabtree, Trans. Soc. Mot. Pict. Eng., XIII (1929), No. 38, 
p. 406. 

"Borax Developer Characteristics," H. W. Moyse and D. R. White, Trans. 
Soc. Mot. Pict. Eng., XIII (1929), No. 38, p. 445. 

"A Quick Test for Determining the Degree of Exhaustion of Developers," 
M. L. Dundon, G. H. Brown, and J. G. Capstaff, /. Soc. Mot. Pict. Eng., XIV 
(April, 1930), No. 4, p. 389. 

"A Replenishing Solution for a Motion Picture Positive Film Developer," 
J. I. Crabtree and C. E. Ives, /. Soc. Mot. Pict. Eng., XV (Nov., 1930), No. 5, 
p. 627. 


"Variation of Photographic Sensitivity with Development Time," R. Davis 
and G. K. Neeland, /. Soc. Mot. Pict. Eng., XVIII (June, 1932), No. 6, p. 742. 

"A Method for the Correct and Most Economical Concentration of Elon and 
Hydroquinone in a Borax Developer," A. M. Gundelfinger, J. Soc. Mot. Pict. 
Eng., XX (April, 1933), No. 4, p. 343. 

"Some Properties of Two Bath Developers for Motion Picture Film," J. I. 
Crabtree, H. Parker, Jr., and H. D. Russell, /. Soc. Mot. Pict. Eng., XI (July, 
1933), No. 1, p. 21. 

"The Deterioration of Sulfite Hydroquinone Solutions and the Mode of Activity 
of Old Solutions," J. Pinnow, Z. Wiss. Phot., 27 (Nov., 1930), No. 11/12, p. 344. 

"The Role of Sulfite in Photographic Developers," J. Rzymkowski, Camera 
(Luzern), 9 (1930), No. 5, p. 128; No. 6, p. 164. 

"Metol Developer," A. Kachelmann, Reproduktion, 3 (May, 1932), No. 5, 
p. 66. 

"Metol-Quinol Developers for Negatives and Prints," G. W. Pritchard, Brit. 
J. Phot., LXXIX (April 22, 1932), No. 3755, p. 240. 

"Advantages of Compensating Developing," A. Lux, Photofreund, XII (Sept. 
20, 1932), No. 18, p. 340; (Oct. 5, 1932), No. 19, p. 357; (Oct. 20, 1932), No. 20, 
p. 377. 

"Practice of Fine Grain Development," K. Brandt, Filmtechnik, VII (Nor. 
14, 1931), p. 7. 

"Deterioration of Amidol Developers," P. J. Cammidge, Amat. Phot., LXXIV 
(Aug. 17, 1932), No. 2284, p. 154. 

"Keeping Properties of Developing Solutions," H. W. Bennett, Brit. J. Phot., 
LXXIX (Sept. 23, 1932), No. 3777, p. 375. 

"Obtaining Sediment Free Developers," E. Van Beugen, Focus, XIX (June 25, 
1932), No. 13, p. 399. 

"Pyrocatechol Developers with Little or No Sulfite," P. Hanneke, Phot. 
Chronik., XXXIX (Oct. 25, 1932), No. 28, p. 209. 

"Motion Picture Developer Formulas," Amer. Cinemat., XIII (Aug., 1932), 
No. 4, p. 42. 

"Reducing Grain in Negative," S. Moir, Amer. Phot., XXVI (Oct., 1932), No. 
10, p. 584. 

"The Grain of the Negative," C. Emmerman, Atelier, XXXIX (Oct., 1932), 
No. 10, p. 82. 

"Perpetual Life Film Tank Developer," W. G. Barker, Brit. J. Phot., LXXX 
(Jan. 13, 1933), No. 3793, p. 16. 

"Fog Formation by Chemical Reactions," E. Fuchs, Z. Wiss. Phot., XXXII 
(1933), No. 1, p. 2. 

"Grain vs. Exposure: Paraphenylenediamine Developer," Brit. J. Phot., 
LXXX (Feb. 10, 1933), No. 3797, p. 69. 

"Keeping Quality of Developers," H. W. Bennett, Brit. J. Phot., LXXIX 
(Dec. 9, 1932), No. 3788, p. 751. 

"Too Little Known Developer," W. Mernsinger-Beat, Camera (Luzern), 10 
(Feb., 1932), No. 8, p. 266. 


"Knapp System of Development," A. Knapp, Brit. J. Phot., LXXX (April, 
7, 1933), No. 3805, p. 191. 

Development Control 

"Actinic Measurements on the Exposing and Printing of Motion Picture 
Film," W. E. Story, Jr., Trans. Soc. Mot. Pict. Eng., XVII (1921), No. 13, p. 106. 

"A New Sensitometer for the Determination of Exposure in Positive Printing," 
L. A. Jones and J. I. Crabtree, Trans. Soc. Mot. Pict. Eng., VI (1922), No. 15, 
p. 89. 

"A Motion Picture Densitometer," J. G. Capstan" and N. B. Green, Trans. 
Soc. Mot. Pict. Eng., VII (1923), No. 17, p. 154. 

"An Improved Sector Wheel for Hurter and Driffield Sensitometry," M. 
Breifer, Trans. Soc. Mot. Pict. Eng., IX (1925), No. 21, p. 85. 

"A Compact Motion Picture Densitometer," J. G. Capstaff and R. A. Purdy, 
Trans. Soc. Mot. Pict. Eng. t XI (1927), No. 31, p. 607. 

"A Trial and Error Method of Preparing a Motion Picture Sensitometer 
Tablet," J. I. Crabtree and C. E. Ives, Trans. Soc. Mot. Pict. Eng., XI (1927), 
No. 32, p. 740. 

"Artificial Sunlight for Photographic Sensitometry," R. Davis and K. S. 
Gibson, Trans. Soc. Mot. Pict. Eng., XII (1928), No. 33, p. 225. 

"The Measurement of Density in Variable Density Sound Film," C. Tuttle 
and J. W. McFarlane, /. Soc. Mot. Pict. Eng., XV (Sept., 1930), No. 3, p. 345. 

"Two Special Sensitometers," D. R. White, /. Soc. Mot. Pict. Eng., XVII 
(March, 1932), No. 3, p. 279. 

"Gamma by Least Squares," D. R. White, /. Soc. Mot. Pict. Eng., XVIII 
(May, 1932), No. 5, p. 584. 

"The Relation between Diffuse and Specular Density," C. Tuttle, /. Soc. 
Mot. Pict. Eng., XX (March, 1933), No. 3, p. 228. 

"Photographic Sensitometry," L. A. Jones, /. Soc. Mot. Pict. Eng., Parti: 
XVII (Oct., 1931), No. 4, p. 491; Part II: XVII (Nov., 1931), No. 5, p. 695; 
Part III: XVIII (Jan., 1932), No. 1, p. 54; Part IV: XVIII (March, 1932), 
No. 3, p. 324. 

"Time and Temperature vs. Test System for Development of Motion Picture 
Negative," W. Leahy, /. Soc. Mot. Pict. Eng., XVIII (May, 1931), No. 5, p. 649. 

"Sensitometric Control in the Processing of Motion Picture Film in Holly- 
wood," E. Huse, /. Soc. Mot. Pict. Eng., XXI (July, 1933), No. 1, p. 54. 

"The Eastman Type 2B Sensitometer as a Control Instrument in the Processing 
of Motion Picture Film," G. A. Chambers and L D. Wratten, /. Soc. Mot. Pict. 
Eng., XXI (Sept., 1933), No. 3, p. 218. 

"The Processing of Variable Density Sound Records," R. F. Nicholson, 
/. Soc. Mot. Pict. Eng., XV (Dec., 1930), No. 6, p. 374. 

"A Motion Picture Laboratory Sensitometer," L. A. Jones, /. Soc. Mot. Pict. 
Eng., XVII (Oct., 1931), No. 4, p. 536. 

"Sound Film Developing and Processing," Kinemat. Weekly, 152 (Nov. 7, 1929), 
No. 1177, p. 52. 

"Making Sound Films (III) Sensitometric Tests," T. T. Baker, Kinemat. 
Weekly. 155 (Jan. 16, 1930), No. 1187, p. 67. 


"Review of American Film Technique," W. Geyer, Kinotechnik, II (Dec. 5, 
1929), No. 23, p. 623. 

"Sensitometric Control in the Development of Sound Films," A. Kuster and 
R. Schmidt Kinotechnik, XIII (April 5, 1931), No. 7, p. 123. 

"Temperature Control During Film Development," T. T. Baker, Kinemat. 
Weekly Supp., 721 (June 18, 1931), No. 1261, p. 41. 

"Saving Underexposures," K. Reitz, Brit J. Phot., LXXIX (Nov. 4, 1932), 
No. 3783, p. 670. 

"After-Treatment of Negatives (IV)," O. Mente, Reproduktion 3 (Nov., 1932), 
No. 11, p. 164. 


PRESIDENT GOLDSMITH : I desire again to stress the need that this Committee 
should give consideration to the new problems that are arising in connection with 
the suitable development and proper maintenance of film for use in extended fre- 
quency systems. 

MR. CRABTREE: The smaller laboratories are often in a quandary as to what 
kind of developing machine to purchase on the market. I wonder whether the 
Committee could make a survey of the available machines and describe them 
in the report. 

MR. HYNDMAN: The Committee did what Mr. Crabtree suggested; but un- 
fortunately, the patent litigation on developing machines has prevented our 
making such suggestions. Furthermore, it is the custom for practically all labo- 
ratories that use continuous machines to build their own; and though the gen- 
eral principles are always the same, many of the adjuncts are designed by the 
members of the particular laboratories. In view of these facts, we did not feel 
that it was advisable for the Committee to recommend what machines should 
be purchased until the litigation will have been completely settled. 

MR. CRABTREE: I don't see that the Committee would have to assume any 
responsibility. It is simply a matter of abstracting the manufacturer's litera- 

PRESIDENT GOLDSMITH: There is no just reason why the publication by the 
Society of descriptions of the various machines that are freely offered for sale by 
their manufacturers would be construed as a violation of legal or ethical rights. 
The buyer may require patent advice if there is litigation in process; but that 
is no reason why the Society should not describe the machines. 

MR. NICHOLSON: I should like to add to Mr. Hyndman's remarks. With 
another member of the Committee I listed the names of all companies that had 
manufactured developing machines, and the types of machines that were being 
used in the various major laboratories, for inclusion in the report. At one of the 
meetings of the Committee, several members objected to including such data, 
so they were omitted. 

MR. CRABTREE: I understand that small machines are manufactured by 
Debrie in France, Geyer in Germany, and Vinten in England. 


The Projection Practice Committee presented its latest detailed 
report at the Spring, 1933, Convention at New York, N. Y. 1 The 
meetings that have been held since that time have been devoted to 
formulating the plans for its next detailed report, to be presented at 
the Spring, 1934, Convention. Some of the topics to be included in 
that report, according to the present plans, are as follows : 

(1) The presentation of the electrical, optical, and mechanical principles in- 
volved in projection, from the point of view of the practical projectionist. 

(2) Projection room routine and maintenance. 

(3) Precautions to prevent breakdowns or imperfections of projection. 

Some of these subjects have already been given considerable atten- 
tion by the Committee ; however, in view of their great importance, 
and in view of the changes of technic that are continually occurring, 
it is advisable that the previous analyses be considerably extended 
and brought up to date. 

Another matter of great importance, to which the Committee has 
already devoted much time and effort, relates to the preparation of 
data dealing with the clearances, tolerances, and amount of wear of 
projection and sound equipment in the theater; and the design and 
use of tools for determining, rapidly and accurately, the condition of 
the equipment of the projection room, and of determining the need 
for replacement of worn parts. The Committee believes that the 
availability of data, tools, and testing methods appropriate to the 
projection room will constitute a contribution of major value to the art 
of projection. 

In concluding this brief report, the Committee wishes again to 
emphasize a statement made in its previous report: "He (the pro- 
jectionist) should be stationed constantly at the projector while it is 
in operation. ..." The Committee is of the opinion that a proper 
factor of safety in operation, as well as the avoidance of imperfect 
operation of the projection equipment or unjustified interruptions 
of service, can be attained only by having an adequate personnel, 

* Presented at the Fall, 1933, Meeting at Chicago, 111. 



as mentioned above, in the projection room. The Committee pur- 
poses further to study this aspect of projection practice, from the 
point of view of technical requirements; and plans to report in full on 
the subject on a later occasion. 

H. RUBIN, Chairman 




J. J. FINN R. H. McCuLLoucH 





1 Report of the Projection Practice Committee, /. Soc. Mot. Pict. Eng., XXI 
(Aug., 1933), No. 2, p. 89. 


MR. CRABTREE : Can the Committee make any recommendations for prevent- 
ing the screens from going dark in the theaters other than by having the pro- 
jectionist keep his eyes glued to the screen, or to the projector? 

MR. GRIFFIN: There are only two reasons for stoppage in a theater, and 
they are film breakage and equipment failure. Unfortunately, these will occur 
sometimes when least expected, but if the projection room is equipped with 
sufficient and proper personnel, there is little excuse for it. 

MR. RICHARDSON: The place for the projectionist is beside the projector 
every minute of the time it is running. If he is not there, there is likelihood that 
fire may occur. Furthermore, the manager says to himself, "Well ! The motor is 
running the show. Why should I pay that fellow $50 a week?" If the men 
stay where they belong and attend to their business, and are given the proper 
cooperation by the management, there is no excuse for a stop. 

PRESIDENT GOLDSMITH: The Committee has not unduly stressed that point. 
One might imagine what might happen if a locomotive engineer should open 
the throttle wide and then retire to the tender to read a novel. It is true that 
a motion picture projector may operate as long as current is supplied to it. 
But that procedure is not in accord with the art of practical projection. This 
Committee of the Society, being a practical projection Committee, is of the be- 
lief that there should be adequately trained and interested personnel in the pro- 
jection room. The Committee is interested only in the idea of satisfactorily 
continuing the show, and of the factor of safety for the audience that is required 
to ensure confidence. 


This Commitee in the past few months has been occupied with 
bringing together biographical records of many of the pioneers of the 
motion picture. These records, which are authenticated by early 
documentary evidence, patents, and contemporary accounts, are to 
be published when complete in the JOURNAL. It is the hope of the 
Committee to create a complete record of the beginnings of the mo- 
tion picture for the JOURNAL so that future students who endeavor to 
establish and to investigate the origins of the industry can find an 
authoritative account in the JOURNAL. The Committee will appreciate 
any suggestions regarding memoirs, documents, or other information 
that will throw light on the early developments. 

Besides this activity of creating written records, the Committee 
has been bringing many accessions to the Los Angeles Museum Col- 
lection of our Society. To enumerate them all would require too 
much space; however, a few of the larger additions will be itemized. 

At last the animated cartoon exhibit has been completed and is on 
display. This exhibit traces the history of the animated cartoon, 
and includes examples of the series of drawings, three feet in length, 
that were used in the Wheel of the Devil in 1834 by William George 
Horner, as well as copies of three-color cartoons made in 1932-33. 
Examples of the cartoons that initiated the cartoon vogue in 1913-15 
may also be seen. There is a complete display of the Disney ' ' Mickey 
Mouse" cartoon covering the process in its various departments by 
photographs. There are examples of the double sound and picture 
manuscript used in cartoon making because the picture and sound 
are recorded separately. There are also examples of the camera ex- 
posure sheets, which show the necessary series of drawings required 
for the action. Stages in the making of the drawing are shown, from 
the rough character drawing layout, to the final opaqued cartoon 
characters. Included are examples of the various types of back- 
ground drawings, original drawings from the first two-color cartoon 
made by Ted Eshbaugh as well as the first three-color "Silly Sym- 
phony," Flowers and Trees, made by Disney. 

* Presented at the Fall, 1833, Meeting at Chicago, 111. 



An exhibit of the various lamps used in the studios has also been 
put on display. In this collection, which was made available by 
the General Electric Co. and Mr. John Winchester of the Metro-Gold- 
wyn-Mayer Studios, are a representative group of lamps used in the 
studios today including the smallest globe made, the lV2-volt "grain 
of wheat" lamp used in miniatures, and the largest lamp, the 10-kilo- 
watt, occasionally used for set illumination. Besides miniature lamps 
and studio set lamps, there are examples of the photoelectric record- 
ing and reproducing cells, projection lamps, heat lamps, etc. As a 
background of this display, there are some historic lamps. One of 
these is a model of the first Edison lamp of 1879, and another is a crude 
globe made by Sawyer in 1879. This last named type was developed 
by many experimenters in electric illumination before Edison con- 
ceived of the necessity of high resistance for efficient illumination. 
There are two coiled platinum wire filaments in the Sawyer lamp 
about I /B inch thick. This collection, however, is far from complete. 
It is required that any one having exhibits either of a historic nature 
or new types of lamps should send examples to the Committee for 
display. They should be forwarded to the Chairman of this Com- 
mittee, care of the Los Angeles Museum, Motion Picture Division, 
Los Angeles, Calif. Besides an official acknowledgment from the 
Museum, credit will be giveii the donor on the accompanying museum 

Mr. L. B. Mayer has presented a collection of pre-sound arc light- 
ing equipment. In this accession are representative examples of the 
"overhead dome," "scoop," "flood banks," "rifle arcs," and "broad 
lights" of the "hard light" period of motion pictures. This collection 
supplements the one loaned some time ago by Mr. O. K. Oleson. It 
is hoped that it will be possible to build a full size set of a historical 
nature showing this lighting equipment, with cameras and other 
paraphernalia, when sufficient space may be alloted to do so. Models 
of the Aristo and early Kleig lights are wanted and any information 
that will lead to the whereabouts of such equipment will be appre- 

Mr. W. Clendenin has added a group of early equipment catalogues 
to his already extensive donation. Mr. C. Blackstone has donated a 
Lubin Camera and Mr. A. J. Fitzpatrick, a model 5 Prestwich 
Camera. There are now in the collection, ten examples of cameras 
that were used during the period of about 1900 to 1915. Mr. Louis 
Lumiere has sent a Cinematographe dating before 1900, which has ar- 


rived but at present is going through the United States Customs rou- 
tine required for historical pieces. This camera is one of the more 
notable of the early cameras. It served as a printer and projector, 
as well as a camera. It is about the size of a cigar-box, while the 
cameras of its time were as large as trunks. Such cameras as the 
Lubin, Pathe, Selig-Shustek, and others patterned their intermittent 
movement after the Lumiere camera, which was first made early in 
1895. It is very important that an old Biograph camera be located 
before they are all destroyed. 

Mr. Cecile B. DeMille has added a number of props used in filming 
the Sign of the Cross. Among them is a replica of the Arena Roll that 
served as programs for the Roman Gladiatorial contests. There is 
a model of a Roman Arena also, as well as other items used by the 
Romans. Mr. L. B. Mayer recently loaned the collection a five-foot 
copper model of a Roman Fighting Galley, which was used in filming 
the naval battle in Ben Hur. 

Three of the proposed set models showing the making of pictures 
today and the historical set have been put on display. These include 
the making of a glass matte shot, a sound set in operation, and a model 
of the first motion picture studio, the Edison "Black Maria," made 
from specifications furnished by Mr. W. K. L. Dickson who supervised 
its construction and made pictures in it for Edison in 1892-95. 

Some additional documents from old magazines and books have 
been located. Where the original was not available, photostats have 
been made. Assistance in locating this material and in photostating 
it is being given by Mr. L. G. Young. The compilation of early refer- 
ences to the industry for posterity and for the use of students today 
is important and is not being overlooked. Anything located should 
be sent to the museum for display and for preservation. Booklets 
and other literature sent out by manufacturers or those developing 
new types of equipment are being filed, and your Committee would 
appreciate anyone's forwarding such data to the museum as it also is 
of value in preserving the history of the industry. Besides the mem- 
oirs of the past, everything pertaining to the present era is being 
preserved. Handbills, technical information, or new apparatus 
announcements are being classified. The importance of this work will 
be recognized by those who have found it necessary to investigate the 
details of almost any activity of the past. 

Electrical Research Products, Inc., has assured the Committee of 
cooperation in bringing together a collection of sound recording 


equipment for the exhibit. Some of this material in the form of 
recording lamps has been placed with the museum. It is their plan 
to display recording slits, microphones, illustrations, and other items 
that will show the sound recording process. 

One of the largest accessions received recently consisted of about 
10,000 old motion picture announcement, lecture, advertisement, and 
song slides, donated by Mr. H. Ross. These slides date back to the 
inception of the use of slides in connection with the motion picture. 
Besides the usual announcements of new pictures and neighborhood 
advertisements, of the time before the trailers on film, there is a 
wide collection of slides exhorting the members of the audience to 
"Please refrain from spitting on the floor," ''Ladies, please remove 
your hats," "Those desiring to smoke, please sit on the right side of 
the house," "Leave the dogs at home," "One minute for repairs," 
"Good night," etc. This collection vividly portrays the boisterous 
movie audience of that day. A representative group has been put on 
display, shown to advantage by back lighting. As a contrast to these 
slides, specimens of the present film trailers may be seen. 

Cooperation is invited and any one knowing of the whereabouts of 
relics or memoirs of the motion picture should get in touch with the 
Chairman of this Committee, care of the Los Angeles Museum, Los 
Angeles, Calif. 

The assistance and help so generously given by Dr. Lee de Forest, 
Robert G. Linderman, Harry Tucker, Willis O'Brien, Carrol Shepp- 
hird, Dr. E. M. Honan, G. R. Lilly, Edward T. Estabrook, Silas Sny- 
der, Walt Disney, Ted Eshbaugh, Walt Lantz, John Winchester, 
Louis B. Mayer, and others has been appreciated. 

E. THEISEN, Chairman 






The last report of the work of this Committee was presented to 
the Society at the New York meeting in April. 1 The present report, 
therefore, is a resume of the subjects discussed and the resulting ac- 
tion since the April Convention. 

Howell and Dubray 2 , in the April, 1932, JOURNAL, proposed a new 
perforation to replace the existing 35-mm. positive and negative film 
perforations. The Standards Committee has discussed at great 
length the question of adopting a single standard perforation for both 
positive and negative film. Briefly, the conclusions are as follows: 
An entirely new perforation would involve new perforating equipment, 
and would not be satisfactory in existing projection equipment. The 
present positive perforation results in longer life of the negatives, 
due to the better clearance when the film is run on commercial sprock- 
ets. Tests made on sound track prints show an advantage in using 
the present positive perforation for both positive and negative film. 
The advantages are better definition of the high frequencies and more 
accurate location of the sound track. It is also possible to print the 
picture with less side-weave if suitable equipment is provided for 

The Standards Committee has unanimously approved the following 
proposal : 

RESOLVED, that a single perforation be adopted for all 35-mm. film and that this 
perforation be the present standard positive perforation, to be known hereafter 
as the Standard S. M. P. E. Perforation. 

It appears that the only equipment that will require a change in 
order to standardize the positive perforation for negative film com- 
pletely will be the registering pins on cameras or printers. Undoubt- 
edly, considerably longer life of negatives will result from using the 
positive type of perforation. 

The Sub-Committee on Sensitometry has agreed upon a standard 
unit of photographic intensity; and the Standards Committee has 
unanimously approved the following resolution. 

* Presented at the Fall, 1933, Meeting at Chicago, 111. 



RESOLVED, that the unit of photographic intensity adopted for negative ma- 
terials by the International Congress of Photography, and the principle of non- 
intermittency in making sensitometric measurements, be adopted as recommended 

The revision of the standards booklet is nearing completion. The 
booklet will contain new drawings for 35-mm. sound film nega- 
tives and positives, showing minor changes in the dimensions and the 
values of the tolerances where desirable. The 35-mm. sprocket draw- 
ings are being revised. The positive type of perforation will be shown 
on the layout for the negative film. The booklet will contain the 16- 
mm. sound-on-film layouts that have been approved. New dimen- 
sions will be shown for the camera and projector apertures, in accor- 
dance with the present practice and previous reports published by the 
Committee and by the Academy of Motion Picture Arts and Sciences. 
The revised booklet will be submitted to the American Standards As- 
sociation after being approved in its final form by the Standards 
Committee, and subsequently published in the JOURNAL and finally 
approved by the Board of Governors. 

The Committee has, in reviewing the standards booklet, given con- 
sideration to comments contained in an article by H. Pander 3 of Berlin, 
entitled "The New German Standards," in which the author compares 
the German standards with those published in the Standards Booklet 
approved by the American Standards Association in September, 

A letter has been received from the British Kinematograph Society, 
stating that at a meeting of the Standards Committee of that body 
held on July 14, 1933, it was agreed that the positive perforation di- 
mensions should be the universal standard for all 35-mm. film. The 
opinion of the S. M. P. E. Standards Committee was also requested 
as to a suitable date for making the change in the negative perfora- 
tion. The date July 1, 1934, was suggested. In the same letter, 
the Chairman of the Standards Committee of the British Kinemato- 
graph Society stated that it had been agreed to recommend a universal 
core for all 35-mm. raw stock, according to drawings transmitted with 
the letter. This suggestion has been discussed by the S. M. P. E. 
Standards Committee and, while no definite action has been taken, it 
is the feeling of the majority of the members of the Committee that 
the suggestion is worthy of further consideration. The British Com- 
mittee also has recommended that a triangular sign be placed between 
film perforations in order to distinguish safety film from nitrate film. 


The reason for such a recommendation is that edge printing as now 
applied to sound film frequently obliterates the present marks along 
the edge of the film. Although the Standards Committee is consider- 
ing this suggestion at the present time, no action has yet been taken. 
The design of a suitable reel for 35-mm. film has been discussed by 
the Standards Committee on several occasions. At the present time, 
the Sub-Committee on Exchange Practice is studying the problems 
involved in connection with reels, and it is expected that this Sub- 
Committee will present a definite recommendation to the Standards 
Committee for a satisfactory reel. 

M. C. BATSEL, Chairman 










1 Report of the Committee on Standards and Nomenclature, /. Soc. Mot. Pict. 
Eng., XX (June, 1933), No. 6, p. 505. 

8 HOWELL, A. S., AND DUBRAY, J. A.: "Proposed Change in the Present 
Standards of 35-Mm. Film Perforations," J. Soc. Mot. Pict. Eng., XVIII (April, 
1933), No. 4, p. 503. 

3 PANDER, H.: "The New German Standards," Filmtechnik, IX (May, 1933), 
No. 8, p. 121. 



On page 395 of the November, 1933, issue of the JOURNAL, the equation im- 
mediately below Fig. 4 should read 

and Eq. 6, on the same page, should read 

[i . / 61 sin o> + 6 2 sin 2w \ 
1(1 -) +*!+{ +.... ) (6) 

+ m { c\ cos w + c<i cos 2o> + . . . . } 



Summary. A resume of the existing situation with respect to sprockets for 
35-mm. sound and projection equipment is first presented, followed by a discussion 
of the interference between the film and the sprockets that transport the film, for 
sprockets of various dimensions. As a result of the study, sets of dimensions are 
proposed for the feed- and hold-back sprockets and for the sound sprocket, submitted 
for the consideration of the Standards Committee of the Society. 

The purpose of this paper is to present to the Society a resum6 of 
the existing situation, and a proposed change in standards, as regards 
35-mm. sprocket dimensions. 

In the Society's standards booklet, ASA-Z22-1930, charts 6 and 7, 
respectively, set forth the standards for feed and hold-back sprockets. 
These standards have frequently been the subject of discussion in the 
Society, and the author personally has never agreed that the dimen- 
sions shown are satisfactory. Since the publication of the standards 
booklet, many new standards have come into existence, and a revision 
of the entire booklet by the Committee on Standards and Nomencla- 
ture has become necessary ; and as the subject of sprocket dimensions 
has again come to the front, the author was requested to prepare for 
the Standards Committee a set of sprocket dimensions that would ful- 
fill the requirements for shrunk and unshrunk film, without damaging 
the film, and would present a minimum amount of interference 
between the film and the sprockets in projection and sound repro- 
ducing equipment. 

All film used in projection and sound reproducing equipment is 
definitely guided at the sound track edge, and sprockets should be 
constructed according to the above requirements with that fact 
clearly in mind. The present standard for feed sprockets as it 

* Presented at the Fall, 1933, Meeting at Chicago, 111. 
** International Projector Corp., New York, N. Y. 



appears in the standards booklet is shown in Fig. 1. It will be 
noted that the transverse tooth gauge is 1.107 inches, the width of 
the sprocket tooth 0.065 inch, the thickness of the tooth 0.050 inch, 
the base diameter 0.9452 inch, and the tooth radius 0.075 inch. It 
will also be noted that the corners of the teeth are shown rounded to an 
approximate radius of 0.010 inch. I wish to point out that if sprock- 


55% FILM 


FIG. 1. Present S. M. P. E. standard feed sprocket. 

ets were manufactured in accordance with those dimensions it would 
be impossible to accommodate standard 35-mm. film properly, when 
using the sound track edge of the film as a fixed guide, without 
seriously damaging the film and sadly marring the quality of the 
reproduced sound. 

The proper design of sprockets, particularly for sound reproduc- 
tion, requires absolute freedom from obstruction between the tooth 



[J. S. M. P. E. 

and the sprocket hole on both sides of the film; and for mechanical 
reasons, the circumferential thickness of the tooth should be as great 
as possible without causing interference between the film and the 
teeth in contact. 

Fig. 2 shows the condition obtained with 35-mm. standard un- 
shrunk film, guided at the edge, and a sprocket having the dimen- 
sions of the Society's present standard, as shown in Fig. 1. For 
reasons that will be explained later, no radius has been shown on 
the corners of the sprocket teeth; and it will be noted that with 
such a condition, adequate clearance obtains in both sprocket holes. 

FIG. 2. Relation between sprocket teeth and film perforations: present 
S. M. P. E. standard sprocket (Fig. 1), and standard 35-mm., unshrunk, 
edge-guided film. 

Fig. 3, however, illustrates the condition that exists when using 
the same sprocket and the same guided edge, when the film, 
has shrunk 1.5 per cent. The sprocket tooth at the picture 
side of the film interferes with the film by 0.0146 inch, and the entire 
strain of moving the film is placed at one corner of this sprocket hole, 
the film being prevented from contacting properly with the driving 
surface of the tooth to the extent of 0.0067 inch. 

Fig. 4 illustrates the situation more forcibly, and shows exactly 
what must occur when the film-driving surface of the sprocket tooth 
is forced into its driving position against the film at the edge of the 
perforation. It is evident that the corner of the sprocket hole must 

Jan., 1934] 



be entirely ripped away. It is also evident that even with a radius 
of 0.010 inch for the corners of the sprocket tooth there would still 
remain an interference of 0.0094 inch at the corner of the perforation 
farthest from the guided edge. 

Referring now to the radius of 0.010 inch for the corner of the 
sprocket tooth, I wish to point out that it is incorrect in at least four 
respects : 

(1) It should never be regarded as a clearance dimension between 
the perforation and the tooth because it disappears as the film wears 
down the tooth; and if the sprocket tooth is not properly designed 


/. 3554-- 

FIG. 3. Same as Fig. 2, but for film shrunk 1.5 per cent. 

to eliminate interference the film will be damaged when the radius is 

(2) If the sprocket tooth is properly designed, so that when the 
radius does wear down there is still no interference between the 
perforation and the sprocket tooth, the working surface of the tooth 
when new is reduced by 0.020 inch, thus providing a tooth using 
only 36 per cent of the available working area of the sprocket per- 

(3) There are 128 such radii on every sixteen-tooth sprocket, and 
it is practically impossible to make them. 

(4) The manufacturing of a sprocket with 128 radii would be- 



[J. S. M. p. E. 

come an intricate problem, indeed, and the cost would be far greater 
than the cost of sprockets as manufactured today. 

The radius is entirely unnecessary in any case, as will be shown, 
and satisfactory sprockets can be manufactured in which the film 
bearing surface is considerably greater than that possible with 
sprockets made according to the Society's present standards. What 
happens to the film when it removes the radius from the sprocket 
tooth of our Society's present standard has been shown in Fig. 4. 

For reasons already mentioned, sprockets constructed according 
to the present standards have, so far as the author has ascertained, 









atff - 

r/^Y &H8UV* 1.5 % 

FIG. 4. Same as Fig. 2; redrawn to show forcing of film at corner of per- 

never been made commercially, and certainly not for use on projec- 
tion and sound reproducing equipment. The International Projector 
Corp. has made hundreds of thousands of sprockets for its own use and 
for the use of other organizations. Feed sprockets manufactured 
until quite recently had a transverse gauge of 1.102 inches (instead 
of 1.107), a transverse width of tooth of 0.060 inch (instead of 0.065), 
a circumferential thickness of tooth of 0.060 inch (instead of 0.050), and 
a tooth radius of approximately 0.085 inch (instead of 0.075). 
More recently, however, the dimensions were changed to those 
shown in Fig. 5. It will be noted in this that the standard base 
diameter of feed sprockets (0.945) was retained, that the transverse 

Jan., 1934] 



pitch of the sprocket teeth is again 1.102, the transverse thickness of 
tooth again 0.060, and the circumferential thickness of tooth now 
0.055. The tooth radius, approximately 0.077, will be explained later. 
Since the Standards Committee began its work of revision, a 
thorough study has been made of the sprocket situation; and it is a 
fact that with a tooth thickness of 0.060 inch there is no circumfer- 
ential interference with either 1.5 per cent shrunk or unshrunk film, 

FIG. 5. Present commercial sprocket. 

when not more than five teeth are engaged. But there is interfer- 
ence in the case of upper feed-sprockets, where the number of teeth 
engaged is eight or nine. Therefore, the circumferential thickness 
should be changed to 0.055 inch for the teeth of all projection and 
sound sprockets. Such a dimension allows satisfactory clearance 
between all sprocket teeth and sprocket holes up to nine in contact, 
with a base diameter of 0.945 inch, and strengthens the tooth by 
0.005 inch, or 10 per cent, when compared with the 0.050 tooth 



[J. S. M. P. E. 

thickness in the Society's present standard. Such a condition is very 
desirable from a manufacturing standpoint. 

In this connection it is quite obvious that the tooth thickness is 
definitely determined by two factors, namely , the base diameter 
and the number of teeth in contact. The base diameter is deter- 
mined by the function of the sprocket, whether feed or hold-back. 
It follows, therefore, that sprockets could be designed with a greater 
tooth thickness for some particular application; but inasmuch as a 
tooth thickness of 0.055 inch and the present standard base diameters 
of 0.945 for feed-sprockets and 0.932 inch for hold-back sprockets 
satisfactorily fulfill all conditions found in projection equipment, 
this tooth thickness should be adopted as standard. 


FIG. 6. Relation between sprocket teeth and film perforations: present 
commercial sprocket (Fig. 5), and standard 35-mm., edge-guided film shrunk 
1.5 per cent. 

The study of the problem also disclosed the fact that with the 
sprocket shown in Fig. 5 no interference occurs with unshrunk 
film, but that with film shrunk 1.5 per cent the condition shown in 
Fig. 6 occurs. The interference there is 0.0097 inch, between the 
corner of the sprocket tooth and the radius of the sprocket hole, 
which keeps the working face of the sprocket away from the film 
by 0.0026 inch. It is evident, therefore, that under such a condition 
either the corner of the sprocket hole is ripped away or the film 
"gives" at the corner sufficiently to allow the face of the tooth to 
come into contact with the driving surface of the perforation. The 
latter is undoubtedly what takes place, inasmuch as there has been 

Jan., 1934] 



no serious evidence of damaged perforations. However, it is not a 
desirable condition by any means, particularly with regard to sound 
sprockets. Note that no radii are shown at the corners of the 
sprocket teeth; the reasons for that have already been mentioned. 
If the radius were present, however, the interference would be 
a clearance of 0.0003 inch. 

During discussions of the subject by members of the Standards 
Committee, it was suggested that it would be generally undesirable 
to change a standard of long standing, and drawings were submitted 

ffV FILM l^lTuOUT GU/Of. 


^goi .qoa^ 



' \ 

\ ^ 

\-M#*\ .<M9fR'\-- 









J~ I 





^ i 



FIG. 7. Relation between sprocket teeth and film perforations: present 
S. M. P. E. standard sprocket (Fig. 1), and standard 35-mm., unshrunk and 
shrunk (1.5 per cent), unguided film. 

for the purpose of showing under what conditions the sprocket shown 
in Fig. 1 could be used satisfactorily. Fig. 7 shows this layout for 
shrunk and unshrunk film, with the Society's present standard 
sprocket drawn in. Note that in this case the film has been shown 
unguided, whereas in practice all film is guided in both the projector 
and the sound reproducer by the edge of the film nearest the sound 
track. If it were possible to manufacture sprockets such as shown 
in this drawing, and unguided film could be used, it would be possible 
to use sprockets such as that which would function satisfactorily 
within the shrinkage limits shown. The drawing shows a radius of 



[J. S. M. P. E. 

0.010 inch at the corners of the sprocket teeth and a working surface 
of 0.045 inch. 

In the layout for unshrunk, unguided film, a clearance is shown 
between the end of the sprocket hole radius and the edge of the 
tooth on the picture side, and an adequate clearance is shown on 
the sound side. In the layout for unguided, 1.5 per cent shrunk 
film, a clearance of 0.005 inch is shown between the end of the 
sprocket hole radius and the end of the sprocket tooth radius (shown 
in dotted lines) on both sides. However, as before stated, it is not 




3 *1 


-."* - o// '-j L j 





J arts'* MS* ( i \ 

\ - \ \ 

1 i 





^ ' 




S~ "- 


FIG. 8. Relation between sprocket teeth and film perforations: present 
S. M. P. E. standard sprocket, with the exception that the teeth are moved 
toward the guided edge by 0.0115 inch, standard 35-mm., unshrunk and 
shrunk (1.5 per cent), edge-guided film. 

commercially practicable to make a 0.010 radius on 128 tooth corners 
per sprocket; and if such a radius is not present, the 0.005 clearance 
shown in the drawing becomes an interference of 0.005, except for 
whatever slight, undefinable reduction might occur from broken 
corners produced in the course of manufacture undefinable, because 
the broken corner is made manually, by filing. However, the fact 
that all film is guided by the sound edge rules out the possibility of 
using such a design. 

Another layout was submitted, for green film and film shrunk 
1.5 per cent, shown in Fig. 8, using the sound track edge of the 

Jan., 1934] 



film as a guide. The sprocket teeth are moved over on the sprocket 
a distance of 0.0115 inch (= 0.1340-0.1225) toward the guided edge 
of the film, which immediately removes it from the category of 
present standards, as shown in Fig. 1. Even were it practicable 
from a manufacturing standpoint (considering again the 128 radii) 
such a sprocket would be impractical from a commercial standpoint 
because it would be impossible to reverse it on the shaft to take 
advantage of the driving surfaces on both sides of the teeth. It is 


FIG. 9. Relation between sprocket teeth and film perforations: proposed 
transverse pitch and tooth width, but with teeth moved toward the guided 
edge by 0.0015 inch. 

the practice of many projectionists to turn sprockets around, thus 
obtaining a 100 per cent increase in the life of the sprocket. 

Referring again to Fig. 8, it will be seen that with green film and 
such a sprocket, if the tooth radius were omitted, the interference 
with the perforation near the guided edge would be 0.008 inch. It 
would be a clearance of 0.002 inch if the radius shown in dotted lines 
were really present. On the picture side, without the tooth radius, 
the interference would be 0.010 inch. With film shrunk 1.5 per 
cent, and without the tooth radius, the interference of 0.0065 shown 
at the guided side would become a clearance of 0.0035 with the 



[J. S. M. P. E. 

tooth radius present as shown in dotted lines. On the picture side 
an interference greater than 0.0031 inch is shown, without the 
radius, which would become a clearance of 0.0068 if the theoretical 
radius were present. As will be shown, the radius is neither necessary 
nor desirable. Such a sprocket as shown in Fig. 8 would require 
the changing of sprocket tension shoes, stripper plates, pad rollers, 
etc., of all the projectors in the field, thus making it impracticable 
from that point of view even if it were satisfactory in all other 

Another layout was submitted using the author's proposed dimen- 

av/oio toct. 

FIG. 10. Relation between sprocket teeth and film perforations: pro- 
posed 35-mm., 16-tooth sprocket, and standard 35-mm., unshrunk, edge- 
guided film. 

sions, given in Fig. 15, except that the tooth thickness was made 
0.050 (instead of 0.55) and again the teeth were moved toward 
the guided edge of the film, this time by only 0.0015 inch. Such a 
layout could be made to work satisfactorily, but it has objections, 
in part, similar to those of the previous layout. The layout is 
shown in Fig. 9; again the sprocket can not be reversed without 
increasing interference between the corner of the sprocket tooth 
and the sprocket hole radius. This sprocket, however, has no tooth 
corner radius, and is more practicable; but it would have to be 
placed on the equipment right end in, and would require an identi- 
fication mark for proper assembly. The worst feature of this sprocket 

Jan., 1934] 



is that if it were reversed, and it surely would be, the clearance of 
0.0015 inch for shrunk film shown for the perforation at the guided 
edge would become an interference of 0.0015 inch, and the inter- 
ference shown for the perforation at the picture edge of 0.0032 
would become an interference of 0.0062, which is a little worse 
condition than can be obtained with film of the same shrinkage and 
proposed standard sprocket, which could be reversed and thus 
permit the same results to be obtained either way. 

We now come to the proposed standard dimensions for 35-mm., 
16-tooth sprockets, which are shown in Figs. 10, 11, and 12. It 

FIG. 11. Relation between sprocket teeth and film perforations: pro- 
posed 35-mm., 16-tooth sprocket, and standard 35-mm., shrunk (1.13 per 
cent), edge-guided film. 

is recommended that the transverse pitch of the sprocket teeth be 
made 1.097 inches, the transverse thickness of the teeth be 0.055 inch 
with the burrs completely removed from the corners, a circum- 
ferential thickness of tooth of 0.055 inch be adopted for all sprockets, 
the center line of the sprocket tooth be 0.139 from either end of the 
sprocket, and that the 0.139 dimension be maintained between the 
center line of the sprocket tooth and the guided edge of the film. 
The tooth radius for the proposed standard is arbitrarily established 
as 0.077 inch, which dimension holds, however, only with respect 
to the upper portion of the tooth; the lower portion is not a true 
radius, but is rather an involute curve developed by the action of 



[J. S. M. P. E. 

the hob. The radius of 0.077 inch is used because it most closely 
approximates the involute curve. 

A sprocket of such dimensions allows a working area of sprocket 
tooth of 0.055 inch as compared with 0.045 inch in the Society's 
present standard, taking into account its theoretical radii. That 
means that the working area of the tooth is definitely increased 
more than 21 per cent for each tooth, and 18 per cent more of the 
available working area of each sprocket hole is used, which is de- 
cidedly advantageous. Fig. 10 shows the proposed form of sprocket 
in operation with unshrunk film. There is a clearance of 0.0025 inch 


AnOtO tC 

FIG. 12. Relation between sprocket teeth and film perforations: pro- 
posed 35-mm., 16-tooth sprocket, and standard 35-mm., shrunk (1.5 per 
cent), edge-guided film. 

at the inside radius of the sprocket hole at the guided edge of the 
film, and a clearance of 0.0015 inch at the inside radius of the per- 
foration at the opposite edge. The transverse pitch is 1.097 inches, 
as mentioned before, and the transverse thickness of the tooth 
0.055 inch. 

Fig. 11 shows the same sprocket operating with film shrunk 1.13 
per cent, no interference whatsoever occurring at either sprocket 
hole. It may be pointed out in this connection that film manu- 
facturers advise that shrinkage of such an order is not present in 
film marketed during the past few years. So much the better; but 
experience shows that we still must allow for transverse shrinkages 

Jan., 1934] 



as great as 1.5 per cent in some of the older product, so Fig. 12 has 
been drawn to show the proposed sprocket operating with film shrunk 
to that extent. It will be noted that there is an interference of 
0.0046 inch at the picture edge sprocket hole, which in effect holds 

FIG. 13. Contour of teeth showing points at which interfering radii 
or fillets occur due to inability of cutting tools to cut absolutely sharp 
corners. At the right is shown the manner of sinking the tooth cutter so as 
to obviate this difficulty; unfortunately, however, burrs and irregular 
surfaces are produced. 

the film away from the working surface of the tooth by 0.0006 inch, 
as compared with an interference of 0.0146 inch between the sprocket 
hole and the sprocket in the case of the Society's present standard 
when the tooth corner radius is omitted. 

FIG. 14. Solution of the difficulties of Fig. 13; the teeth rise out 
of a relieved section, bringing all burrs or fillets below the base diame- 
ter of the sprocket. 

In former methods of manufacture it has been extremely difficult, 
almost impossible, to avoid a small radius or fillet at the base of 
the tooth where it meets the body of the sprocket because the cutting 
tools can not have absolutely sharp corners (Fig. 13). As that is 



[J.'S. M. P. E. 

the working point of the sprocket, interference frequently occurs, 
which prevents the film from seating itself perfectly on the base 
diameter. Such a condition may be partially obviated by sinking 
the tooth cutter slightly below the base diameter of the sprocket, 
which practice, however, unfortunately introduces other evils, 
principally an irregular surface for the film to ride on, with burrs 
which can be completely eliminated only with difficulty. An ex- 

t.cts ours/of &* 'o* fteosFHooir 
/. Oizdvrs/eeoM m* KOID gxtSfRoorgr 

FIG. 15. Drawing of feed and hold-back sprockets 
proposed for standardization. 

cellent solution of this problem is found in Fig. 14, where the teeth 
rise out of a relieved section, thus bringing all burrs or fillets below 
the base diameter of the sprocket, maintaining a smooth even surface 
for the film to ride on. There is nothing on this tooth that might 
cause interference with the perforation, and the film seats itself on 
the base diameter as it should. 

All the foregoing material has dealt entirely with feed-sprockets 

Jan., 1934] 



for projection equipment and hold-back sprockets for projection 
and sound equipment, and a composite dimensional chart of such 
sprockets is shown in Fig. 15. It is quite evident that the proposed 




.943 M*. 
1. 5% SHKl*KH6 


FIG. 16. Relation between sprocket teeth and film perfora- 
tions: sprocket of Fig. 15 used as sound sprocket; base diameter 
0.945 inch, and standard 35-mm. film shrunk 0.15, 0.60, and 1.5 
per cent. 

standard feed-sprocket can be readily used as a sound sprocket also, 
but the problem of correct design for sound reproducer sprockets 
becomes slightly different in so far as the base diameter of the sprocket 





~J\ A 


*w7 *7t* 

.949 DM. 



.942 DIR. 
.675Z SHRINK (Kf. 





.943 J>lf). 

S % 3H&M<fKi. 

FIG. 17. 

Same as Fig. 16. except that the base diameter has been decreased 
to 0.942 inch. 

is concerned. The reason is that it is extremely desirable that the 
sound sprocket transport the film past the scanning beam at as 
constant a velocity as possible; and, therefore, with a minimum 



[J. S. M. P. E. 

amount of slippage from tooth to tooth. A sprocket for that pur- 
pose having a base diameter of 0.945 will perform as shown in Fig. 16, 
in which it will be seen that for a shrinkage of 0.15 per cent all teeth 
are engaged; that with a shrinkage of 0.6 per cent the left-hand 
tooth shown in the drawing does all the work of moving the film; 
and that there is a slippage from tooth to tooth of 0.00084 inch. 
With a shrinkage of 1.5 per cent the left-hand tooth is still doing 
all the work of moving the film, and there is slippage from tooth to 
tooth in this case of 0.0025 inch. 



spzotxerz?/* wwwes. 








.000 28 







.OOO O4 









.OOO 38 







.OOO? 9 





.000 es 









. OOOO9 


.000 3 / 




.OOO 48 

.00 029 









.OOOO 6 











.OO2 3 3 





FIG. 18. Chart showing interference and slippage in re- 
lation to base diameter of feed sprocket, and shrinkage of film. 
Figures above heavy line indicate interference in inches between 
edge of perforation and entering tooth. Figures below heavy 
line indicate slippage in inches from leaving to entering tooth. 

Recent investigation indicates that the average shrinkage of film 
as received for a first showing is between 0.2 and 0.3 per cent, and 
that during the life of the print as a first run subject the shrinkage 
rapidly increases to about 0.5 to 0.7 per cent, and more slowly there- 
after. Theoretically, then, it might be expected that better results 
could be obtained, as regards uniformity of film propulsion, with a 
sprocket having a tooth pitch more closely registering with film in 
this shrinkage range; and that, therefore, a sprocket having a base 
diameter of 0.942 inch would be better for the purpose. Such a 

Jan., 1934] 



sprocket is shown in Fig. 17. It will be noted that with film shrink- 
age somewhere between 0.5 and 0.6 per cent all teeth are in contact 
with the film perforations, which is an ideal state of affairs. It is 
true that when prints are first released the shrinkage is not so high, 
and is of the order of 0.2 to 0.3 per cent, or even lower. The upper 
drawing in Fig. 17 shows the conditions under which the film and 

FIG. 19. Drawing of sound sprocket proposed 
for standardization. 

the sprocket are then working. One tooth is doing the work of 
moving the film, but the following or oncoming tooth is subjected 
to an interference of 0.00062 inch for a shrinkage of 0.13 per cent; 
or, instead of having a slippage from tooth to tooth, the oncoming 
tooth is pushed slightly, this action continuing, in a diminishing degree, 
until a value of shrinkage between 0.5 and 0.6 per cent is attained, 
when all teeth are in contact. As shrinkage increases, a gradual 

38 H. GRIFFIN [j. S. M. P. E. 

increase of slippage from tooth to tooth occurs until, at 1.5 per cent 
shrinkage, the slippage is of the order of 0.0019 inch. In this con- 
nection the sprocket shown in Fig. 17 is superior to a sprocket 
having a base diameter of 0.945 inch. 

Therefore, the proposed standard base diameter for sound 
sprockets is 0.942 inch. This will provide a sprocket that will give 
excellent results under average conditions and satisfactory results 
with film shrinkages between 0.6 and 1.5 per cent. The slight inter- 
ference with the oncoming tooth, as shown in Fig. 18, for shrinkages 
below 0.6 per cent is practically negligible. A careful study of 
Fig. 18 will support this recommendation, notwithstanding the fact 
that the chart also discloses the fact that a sprocket having a base 
diameter of 0.940 provides the best condition for film shrunk 0.675 
per cent. By adopting a sprocket having a base diameter of 0.942 
we would have a sprocket that more satisfactorily accommodates 
shrinkages from 0.15 to 0.675 per cent than does a sprocket having 
a base diameter of 0.940 inch, inasmuch as the slip at 0.6 per cent 
with the 0.942 -inch sprocket is only two and one-half ten-thousandths 
of an inch, and the interference at 0.15 per cent, which is rarely en- 
countered outside the studio and laboratory, is less than six ten- 
thousandths of an inch; whereas, with a 0.940-inch sprocket there 
is an interference of one and one-half ten-thousandths at 0.6 per 
cent and an interference of nearly one thousandth of an inch at 
0.15 per cent shrinkage. The dimensions of the proposed standard 
sound sprocket are shown in Fig. 19. It should be evident from a 
study of the drawings shown in this paper that sprockets of the 
dimensions proposed as a standard offer a satisfactory solution of 
the sprocket dimension problem; it is hoped that they will meet 
with the approval of the Society. 


MR. JONES: How long, under usual conditions, do the sprockets last before 
they are reversed? 

MR. GRIFFIN: That is very difficult to answer, because a projectionist usually 
uses his own judgment. If the sprocket is designed properly and is of the proper 
material, the tooth will wear uniformly from the base to the top of the tooth, 
and as long as there is no under-cutting the sprocket is all right. If under- 
cutting occurs, the sprocket should be changed immediately. Even a very little 
under-cutting is disastrous. It is very hard to say when they should be reversed, 
because that depends on how the tooth performs. 

MR. RICHARDSON: Referring to Fig. 6, at the time of its adoption by the 
Standards Committee, there was much discussion as to whether the corners 


should be rounded or not. It was finally decided that the rounded sprocket 
hole was necessary, in order to increase the strength and to prevent splitting at 
the corners. Projectionists were greatly troubled with split sprocket hole 
corners. Tension in the projector is tremendously variable; many projectors 
used to have, and to a certain extent yet have, excessive tension. If the tension 
is correct and the film is green, a deposit of emulsion accumulates on the aperture 
plate: the tension is immediately greatly increased. Removing the rounded 
corner of the perforation will decrease the resistance of the corner against the 
strain of which I am speaking. Split sprocket holes are bad because they cause 
unsteadiness of the picture on the screen. 

MR. GRIFFIN: The only bad condition that exists (referring to Fig. 6) is 
that the film has shrunk 1.5 per cent; such great shrinkage is not encountered 
in the film that is being manufactured today, but was prevalent in the film that 
was used three or four years ago. 

The sprocket tooth is held away from the film a little more than 0.0026 inch, 
and inasmuch as we have had far greater interference in the past and, as stated in 
the paper, there has been no apparent damage to film, I don't think that is a 
serious thing today. Fig. 4, with the present standard, shows an interference 
of 0.0146 inch. In Fig. 11, of the proposed studies, we have a shrinkage of 
over 1 per cent, which is considered to be excessive today. Even in that condi- 
tion, you see, there is a clearance over both teeth and both sprocket holes. 

MR. CARVER : The chief argument for the reduction in transverse pitch is very 
closely related to edge-guiding. Now, the edge on all the drawings is assumed to 
be 0.139 inch from the center of the sprocket teeth. In actual practice, at least 
until two years ago, the edge guiding has been a very haphazard thing. All the 
edge-guiding in the picture gate was done by the two little rollers at the top. Al- 
though these may have been adjusted correctly at the factory, in practically 
every case the projectionist himself changed their position so that the edge- 
guiding on no two machines would be exactly alike. That means that there 
would be interference in some of the machines on one side, and in other machines 
on the other side. The adjustment would be changed only when the picture 
on the screen was seen to lift up in one corner. Then it would be moved over a 
bit to avoid this trouble. For these projectors the transverse pitch of the sprocket 
ought to be the same as that of most of the film that is being run. Film of which 
the average transverse pitch has shrunk, say, 0.5 per cent, will have a better 
chance of not causing interference on these machines that are out of adjustment 
if the sprockets have the same transverse pitch as the film, than if the transverse 
pitch of the sprockets were as short as is indicated on the drawings, i.e., to cor- 
respond to a shrinkage of about 1 per cent. The real reason why I believe it is 
probably necessary, at least for the International Projector Corp., to make 
sprockets as outlined is that for the last two years a certain number of projectors 
have been put out with definite edge-guiding in the picture gate. That edge 
distance is, as you have shown, 0.139 inch. 

This brings up the question as to whether we should adopt a standard that is 
correct for one projector and perhaps satisfactory, but theoretically somewhat 
incorrect, for projectors that have been designed a little differently. We hope 
that our standards will be adopted and in agreement with the German 
and the International standards. It seems a bit ridiculous to show, as in Fig. 10, 

40 H. GRIFFIN [j. s. M. p. E. 

edge-guiding in which the center of the sprocket on the right-hand side is actually 
nearer the center by 0.0055 inch than the center of the hole. The tolerance for 
shrinkage is reduced just by that amount. If it were moved farther to the right 
there would be much more tolerance. There would be used, in that case, a 
greater transverse pitch, which would be more satisfactory for most of the Simplex 
machines on the market now, in which the edge-guiding is a rather haphazard 
arrangement, the adjustment being made solely by the projectionist. From Mr. 
Griffin's point of view and the point of view of the International Projector 
Corp., I don't see what else can be done. I think probably the thing for them 
to do is to make the sprockets as described. It is quite another thing, however, 
for the Society to sanction a standard that really is illogical. It is probably 
satisfactory, but should we officially adopt such a standard? 

In regard to the rounded corners, Mr. Griffin has presented his point very 
strongly and very clearly against the rounded corners. I must admit that he 
has convinced me. The film manufacturer looks at such things somewhat differ- 
ently from the projectionist. He hates to see all the film that he has so carefully 
made subjected to sharp corners of any sort. We have made some tests with the 
square-cornered type and the rounded-cornered type, and as you all probably 
would predict, found the film to have a life three times as long with the rounded 
corners as with the squared corners. Mr. Griffin's argument that the rounded 
corners don't last, that most of the time the corners are sharp because the sprockets 
are worn, is certainly of great weight; but it seems to me that in an official stand- 
ard, at least, the corners should be rounded a little. 

In regard to the increased bearing surface, a year or so ago, when wide film was 
being discussed, we measured the resistance of the film to the impact of teeth having 
different bearing surfaces, and found that the resistance increased linearly as the 
bearing surface was increased, until it reached a value of about 0.0045 inch. After 
that it didn't matter very much whether the bearing surface was increased or not. 
Of course, a narrow bearing surface such as a knife would cut right through. 
As the bearing surface is increased, the knife becomes duller and duller ; but after 
a while a certain point is reached beyond which further increase in bearing surface, 
or dullness, causes no further increase in resistance. 

MR. GRIFFIN: I stated before that the edge-guiding may be adjusted by the 
projectionist. I should have stated that the adjustment is made on the unguided 
side. The guided side is fixed in the film trap. The adjustment for the tension 
is on the other side, so that we have definite edge-guiding even in the old pro- 
jectors. The new ones, as I explained, have runways to guide the film definitely. 
I am quite sure that sprockets of the design I have indicated wouldn't cause any 
hardship on the old equipment, no matter how old it is. 

MR. CARVER: I mentioned the fixing of the edge-guiding so that the sprocket 
was at the inside of the hole rather than the outside. 

MR. GRIFFIN: The illustration to which you referred was for unshrunk film; 
unshrunk film is unknown, particularly when it is projected in a theater, and 
the condition improves as it shrinks. 

MR. CARVER: No, it gets worse as it shrinks. 

MR. GRIFFIN: Yes, it moves to the right. I get your point, but there is 
a clearance of only 0.0015 inch with unshrunk film. 

MR. CARVER: I admit it is a minor detail. I think this sprocket is much 


better than the old sprocket actually manufactured. I don't know that it is 
better than the old standard, although I think it probably is. 

MR. GRIFFIN: I will say this: the old standard was perfect, without edge- 
guiding. The drawings prove that. 

MR. CARVER: This sprocket is so much better than the sprocket now manu- 
factured that perhaps wear and tear troubles will vanish. The change of diameter 
alone should increase the life of the film five times at least. 

MR. JONES: Why was the value 0.139 chosen? That is the distance between 
the edge of the guide and the center line of the adjacent tooth. From an in- 
spection of the illustration, I believe that if the dimension had been made some- 
what less the shrinkage range would have been considerably greater before the 
beginning of interference on the other side where it does occur. Why did you 
choose 0.139 rather than a slightly smaller dimension? 

MR. GRIFFIN: There is only one real reason, and it isn't a hard and fast rule: 
0.139 happens to be the distance from the center line of the sprocket tooth to the 
end of the sprocket on both sides, giving a sprocket of 1.375 inches wide; it is 
barely possible that equipment might be manufactured in which the guiding 
edge is the edge of the sprocket itself. Under such circumstances that dimension 
would be necessary. The sprocket is placed on the shaft, and the bearings in 
which the sprocket shafts run are milled to an accurate dimension, the width 
of the sprocket definitely placing them at a fixed point with relation to all the 
other sprockets in the projector. We started out with a 1.375-inch sprocket 
some years ago, and that dimension must be retained for the sake of interchange- 

MR. JONES: As a matter of fact, the way edge-guiding is practiced today, 
there is no reason why the 0.139 dimension couldn't be decreased to give you the 
additional shrinkage range. 

MR. GRIFFIN: A study of Figs. 10 and 11 shows that it would be possible to 
move the teeth over only 0.0015 inch, and then only considering that the sprocket 
may not be reversed on its shaft. If it were reversed, and it surely would be, 
interference would occur at a shrinkage of 1.13 per cent. It would not be good 
practice at this late date to begin making sprockets that could not be placed 
on the shafts without regard to which end was placed on first. To move the film 
itself the one and one-half thousandths would require operations on all projectors 
in use. After all, 0.0015 inch is not a serious matter. I doubt whether the 
film path in many projectors would be as accurate as that. 



Summary. A d-c., high-intensity arc for low currents is described. It differs 
from the ordinary high-intensity arc used for projection in that the positive carbon 
is not rotated and is copper coated so that it can be held some distance from the crater 
end. The current, voltage, and burning characteristics are given for the size carbons 
available at this time. 

The possible utilization of this type of arc for projection is discussed, and it is 
pointed out that because of the comparatively simple lamp mechanism required, it 
will take its place with the new a-c., high-intensity arc described in a previous paper 
in supplying the need for more light in the small and medium size theaters. 

In a previous paper 1 the authors have shown that projection lamps 
and carbons may be classified in definite types, depending upon 
the kind of arc, the current, and the optical system used. These 
types are as follows : 

(1) The high- intensity condenser type, with rotating positive, high-intensity, 
d-c. carbons burning at 85 to 150 amperes. These lamps consume approximately 
15 per cent of the carbons used for motion picture projection. 

(2) The high-intensity mirror type, with rotating, positive carbons burning 
at 60 to 85 amperes. These lamps consume about 18 per cent of the carbons 
used for motion picture projection. 

(3) The low-intensity mirror type, with non-rotating, positive, low-intensity, 
d-c. carbons burning at 16 to 42 amperes. These lamps consume about 60 per 
cent of the carbons used in motion picture projection. 

It was also shown that a rather large interval existed between 
the screen illuminations practicable with the low-intensity reflecting 
arcs and the lowest screen illumination obtained with the high- 
intensity reflecting arc. There is also a marked difference in the 
color of the light from these two types of arc. The low-intensity 
arc gives a yellowish white light, whereas the high-intensity arc 
gives a snow white light which is generally considered more desirable 
for the projection of motion pictures. 

* Presented at the Fall, 1933, Meeting at Chicago, 111. 
** Research Laboratories, National Carbon Co., Cleveland, Ohio. 



There was also described a fourth type, utilizing a high-intensity, 
a-c. arc preferably operated on the secondary of a specially designed 
transformer, without the usual motor-generator set and ballast 
resistance. This high-intensity, a-c. arc gives screen illuminations 
higher than the low-intensity arcs, therefore bridging, in part, the 
wide gap between the screen illuminations produced by the low- 
intensity and the d-c., high-intensity reflecting arcs, and makes 



^- E 

3 ~*. 







S N 


















^ s - 

S N 












/ / 




* ^ 







FIG. 1. Intrinsic brilliancy across crater face. 


Positive Negative Amps. Volts 

X 12-Mm. S.R.A. 8-Mm. S.R.A. 30 55 

O 6-Mm. H.I. cc. 5-Mm. cc. 40 31 

D 7-Mm. H.I. cc. 6-Mm. cc. 50 33 

A 8-Mm. a-c. cc. a-c. 80 25 

available to the smaller theaters a white light very similar to that 
obtained from the d-c., high-intensity arcs. 

In this paper will be described another type of light source, which 
may take its place with the a-c. arc in bridging this gap between 
the low-intensity and present high-intensity arcs. 

It is for use on direct current, and employs a series of high-intensity 
carbons smaller in diameter and lower in current-carrying capacity 
than have hitherto been used for projection work. The carbons 

44 D. B. JOY AND A. C. DOWNES [J. S. M. P. E. 

have been designed so that with a few simple precautions the positive 
carbon does not need to be rotated and the negative carbon can be 
coaxial with the positive. The carbons are protected from oxidation, 
and their electrical resistance is reduced by a copper coat that 
makes it practicable to hold the carbons any convenient distance 
from the arc. 

The carbons are at present available in the sizes and current 
ranges given in Table I. Although they are copper coated, smaller 
in size, and of lower current capacity than the 9 -mm. to 16-mm. 
high-intensity carbons used for motion picture projection, these 
carbons have the same crater appearance and formation, high 
current density, and give the same brilliant snow-white light on 
the screen as the larger uncoated, high-intensity, d-c. carbons. The 
rate of carbon consumption is essentially the same as that for the 
larger sizes of high-intensity carbons. 

Table II gives the average intrinsic brilliancy of these arcs and 
other light sources used for projection, and Fig. 1 shows the distribu- 
tion of intrinsic brilliancy of the new light sources. The method 
of making these measurements is the same as that employed by 
Benford and ourselves, and has been described in the literature. 2 ' 3 

While Table II shows that the intrinsic brilliancies of the 6-mm. 
and 7-mm. carbons at the specified currents are approximately the 
same, it should be noted that the crater area of the 7-mm. carbon 
at 50 amperes is greater than the crater area of the 6-mm. at 40 
amperes, and that the 7-mm. carbon gives 44 per cent more light 
than the 6-mm. size at the respective currents. 

It is evident from Fig. 1 and Table II that the crater diameters 
of these non-rotating, high-intensity carbons are considerably smaller 
than those of the 12-mm. mirror arc carbons used in a majority of 
the low-intensity mirror arc applications. This difference in crater 
size will require an optical system of higher magnification when 
using a 7-mm., high-intensity, d-c., positive carbon than one for use 
with the 12-mm. mirror arc carbon to give the same size spot on the 
aperture plate. 

Fig. 1 shows very marked differences in the distribution of light 
intensity across the faces of the various light sources. The crater 
face of the low-intensity, positive, mirror arc carbon is comparatively 
flat. The light comes entirely from the incandescent crater, and is 
therefore fairly uniform across the crater face. 

The crater of the new, non-rotating, d-c., high-intensity, positive 

Jan., 1934] 
































B G 




















^ 00 00 


C., High-Intensity Arcs, Positive Carbon Non-Rotatin 

Current-Carrying Consumption in 
Capacity Inches per Hour 
(Amperes) (Pos.) (Ne 

=U i 

<N <N <M 


iO O O 

O5 OJ 00 

O O O C 

n 22 

CO T^ O t^ 


Brilliancy of the New A res and Other Light Sources 

Candle Power Crater Depth Crater Diameter 
per Sq. Mm. (Inches) (Inches) 










(N (N 








HH HH & 






Negative G 

W K > .& 

a 5 1 

a s ^s 

10 co ^ 






tensity car 

es, 55 volti 
flame carb 

s, 25 volts, 

s, 31 volts. 
i, 33 volts. 







v bo 

K 3 


Positive Carbon 


HH' HH 05 g . 

H4 HH ' CJ 

| | * Q 1 

CD |>. ^ 00 

Light Source 

sten filament 40- watt vacui 

ir blub incandescent lamp, 
sten filament 900-watt gas-1 

ir bulb incandescent lamp. 
ive crater of cored d-c. low-i 

12-mm. carbon at 30 amp 
:r of high-intensity a-c. whil 

8-mm. carbon at 80 ampe; 
ive crater of non-rotating hi 

. white flame carbon. 
6-mm. carbon at 40 ampei 
7-mm carbon at 50 amper 




y . 





S W) 

CJ ~~ 



-* ' ^ 



*y *c 



" O 


46 D. B. JOY AND A. C. DOWNES [J. S. M. P. E. 

carbon is cup-shaped, and is essentially the same as that of the large, 
d-c., high-intensity carbons. The light intensity is considerably 
higher at the center of the crater than at the sides as in all d-c., 
high-intensity arcs, due to the crater shape and the presence of the 
highly luminous gases in the crater. It is essentially the same in 
form, although lower in value, as that found by Benford 3 for the 
150-ampere, high-intensity carbon. 

The a-c., high-intensity carbon has a comparatively flat face, and 
also the highly luminous gases similar to those found in the crater 
of the high-intensity, d-c. carbon. The shape of the distribution 
curve of intrinsic brilliancy is, therefore, intermediate between those 

FIG. 2. Light distribution from arc in horizontal plane. 

Carbons Arc Length 

Positive Negative Amps. Volts (Inch) 

X 12-Mm. S.R.A. 8-Mm. S.R.A. 30 55 11/32 

O 6-Mm. H.I. cc. 5-Mm. H.I. cc. 40 31 5/32 

D 7-Mm. H.I. cc. 6-Mm. H.I. cc. 50 34 7/32 

of the low-intensity mirror arc and the d-c., high-intensity carbon arc. 
These differences in distribution of the light intensity of these light 
sources must be taken into consideration in designing optical 
systems to give the uniformity of screen light required by good 
projection practice. 

The light emitted by the 12-mm., low-intensity arc and the non- 
rotating, d-c., high-intensity, arcs, in a plane through the axis of the 
carbons, is given in Fig. 2. The dip at degrees, or directly in front 
of the positive carbon, is due to the interference of the negative 
carbon and negative holder. The light cut off by the negative carbon 

Jan., 1934] 



3|| o 

S-fiS o 

in A 


5>l S 
| SS 









CO Tf t s * 




M i-J |T] h^ 

i i ^ >- 

a a E s 

E E E E 

o >o co ?o 

a w w w 


i ( 00 ?O CO S- I s - 


o o 


CJ O O 00 O 
,_J H-i M * * 

48 D. B. JOY AND A. C. DOWNES [J. S. M. p. E. 

and holder is in reality only a small percentage of the useful light 
from the arc. 

The non-rotating, d-c., high-intensity arcs are operated at a com- 
paratively low voltage and short arc length, as shown in the figure. 
The light at 90 degrees, or at right angles to the trim, is much higher 
for the high-intensity arc than for the low-intensity arc. Part of 
this light is usable since it includes at least a portion of the focusable 
light directly in front of the crater. These distribution curves may 
be valuable in determining the proper angle of light pick-up for the 
mirrors to be used with these carbons. The mirror size will also be 
governed by the speed of the projection lens, the physical limitations 
of the lamp, and the cost of the mirror. 

When the positive and negative carbons of the non-rotating, 
high-intensity arc are on the same axis, there is no strong directional 
force guiding the tail flame from the arc. It may be desirable to 
use a magnet to attract the tail flame in one direction, although it 
has been found that when a comparatively short arc length such as 
given in Fig. 2 is used, the arc gives steady operation and light 
without superimposing any external force. 

Operation of this non-rotating, d-c. arc with this short arc length 
and low voltage will need a motor-generator set of a rating not 
greater than 50 to 55 volts, and it may be possible to design generators 
of such characteristics that no resistance will be required in series 
with the arc. It should be emphasized that freedom from stray 
magnetic fields and the proper alignment of the negative with re- 
spect to the positive are essential for uniform light and steadiness. 

Table III gives the current, arc voltage, line power, and relative 
light on the screen through the same optical system for the 12-mm., 
S.R.A., low-intensity mirror arc, the 8-mm., high-intensity, a-c. arc, 
and the new non-rotating, high-intensity, d-c. arc. 

These data on comparative light on the screen were obtained on a 
laboratory set-up and, while comparable among themselves, are not 
indicative of what may be obtained with any other set-up of optical 
system and screen. In order to compensate partially for this condi- 
tion the measurements are made with the same distribution of light 
on the screen for all the carbons. 

While Table III shows that at the particular wattages given, the 
non-rotating, d-c., high-intensity arcs give a higher screen illumination 
than the a-c., high-intensity arc of equivalent wattage, it should be 
recognized that the exact order in which these will come in practice 

Jan., 1934] D.-C. NON-ROTATING CARBONS 49 

will be dependent upon the design of the optical systems. These 
optical systems must take into consideration the differences in diam- 
eter, and the shapes of the intrinsic brilliancy curves of the light 
sources if they are to give the same uniformity of screen illumination 
and the same latitude of operation. 

These measurements do not illustrate the difference in the color 
of the light from these types of carbons. The light from the low- 
intensity arc is a brilliant white light, although it appears to be 
yellowish white when compared on a screen with the snow-white 
light given by the a-c., high-intensity arc or that from any of the d-c., 
high-intensity arcs. This is best illustrated by throwing the light 
from a 12-mm. S.R.A. carbon and the new d-c., non-rotating, high- 
intensity carbon on the screen through a suitable optical system. 

It is apparent, from the data presented here, that the conven- 
tional light sources, with the addition of the a-c., high-intensity 
arc and the new non-rotating, d-c., high-intensity arc, provide a series 
of light sources which with suitable optical systems will give light 
on the screen ranging from that desired by the smallest theaters to 
that demanded by the largest theaters. 


1 JOY, D. B., AND DOWNES, A. C. : "A New Alternating- Current Projection 
Arc," /. Soc. Mot. Pict. Eng., XXI (Aug., 1933), No. 2, p. 116. 

2 JOY, D. B., AND DOWNES, A. C.: "Properties of Low-Intensity Reflecting 
Arc Projector Carbons," J. Soc. Mot. Pict. Eng., XVI (June, 1931), No. 6, p. 684. 

3 BENFORD, F.: "The High-Intensity Arc," Trans. Soc. Mot. Pict. Eng., 
IX (1925), No. 24, p. 71. 

4 JOY, D. B., AND DOWNES, A. C.: "Characteristics of High-Intensity Arcs," 
/. Soc. Mot. Pict. Eng., XIV (March, 1930), No. 3, p. 291. 


MR. GREENE : Where, in this scale of light units, would the 75-ampere, rotating, 
high-intensity arc fall? 

MR. DOWNES: Just above the 9-mm. mirror reflecting high-intensity and 
below the 11-mm. at 90 amperes. 

MR. GREENE: How does the amount of light delivered to the screen by the 
arc in its mounting compare with that delivered by a mirror as used in theaters? 
What value would you assign for the column headed "Screen Light in Arbitrary 
Units" (Table III)? 

MR. DOWNES: I assume that you mean a high-intensity arc with a condensing 
lens system, as compared with the reflecting, high-intensity arc. Since the high- 
intensity arcs used with condensing lenses are all burned at higher currents than 
those used with the reflecting mirrors, they would give more light on the screen 


than any of the arcs described in the paper, assuming the optical systems to be 
the same. It should be recognized that differences in optical systems carbon 
trimming, etc. sometimes even in the same theater, cause a great deal of trouble 
in attempts to compare screen illuminations from different carbons. 

MR. MACOMBER: Were the optical systems used here designed primarily 
for the smaller light sources or for larger ones? 

Mr. DOWNES: The same optical system was used in all these tests, adjusting 
the distribution of the light on the screen so that it was the same in each case. 
Optical systems particularly adapted to each of the light sources described in the 
paper were not available. 

The paper carefully points out that when practical lamps and optical systems 
are available, these several light sources may not fall into the exact order shown in 
Table III. 

MR. MACOMBER: Any given optical system would be more favorable to one 
carbon size than to another; it is probable, then, that these were a compromise. 

MR. MARR: What is the effect of these new carbons on colored film? I have 
the impression that the very white light would make colored pictures appear some- 
what hard and cold, and possibly less pleasing than would a somewhat softer light 
of the same intensity. Will this new development be extended to the high- 
powered carbons for searchlight work? 

MR. DOWNES: Spectral energy distribution curves for high-intensity arcs at 
various currents and for several carbon sizes have been previously published by 
the authors in the JOURNAL. 4 We have not as yet made energy distribution curves 
on these new d-c. carbons, but from our work on various types of arcs we know 
quite well that the form of the spectral energy distribution curves will be the same 
as those shown in the article cited, but at a lower level. The effect of these light 
sources with colored film would therefore be similar to that of the high -intensity 

In regard to searchlight applications, we have operated for a long time high- 
intensity searchlight carbons that have been in regular use at various currents up 
to 150 amperes. Experimentally we have made carbons, called super-high- 
intensity carbons, for currents up to 250 amperes. A few such carbons are being 
used regularly at 180 to 200 amperes. Tests on super-high-intensity carbons show 
that the light on a screen located 2700 feet from the searchlight is about 60 per cent 
more at 180-190 amperes, and about 100 per cent more at 250 amperes than that 
from a regular 150-ampere carbon. 


P. MOLE** 

Summary. A new motion picture arc lamp designed for use as a general lighting 
unit is described, in which the new 8-mm. t copper-coated, cored carbons are used. 
Two special mechanisms feed the two carbons of the unit independently of each other, 
the rate of feed of each being controlled by the voltage drop in each arc. Each control 
circuit includes a voltage coil and a current coil, acting in magnetic opposition, 
which arrangement avoids variations of light intensity, flickering, and blinking. 
The lamps operate under a voltage of 115, drawing a current of 40 amperes, d-c. 

During the past five years the development of carbon arc lighting 
equipment for use in motion picture production was retarded by 
several factors. The introduction of panchromatic film, and its 
almost universal acceptance as negative raw stock, provided a 
photographic medium that was well adapted to photographing 
with incandescent filament lamps. The introduction of sound re- 
cording in connection with motion picture photography prohibited 
the use of any type of lighting equipment that was not quiet in opera- 
tion. However, all through this period 150-ampere Sun arcs and 
Rotary Spots have been used to a large extent in combination with 
incandescent lighting. 

By making mechanical improvements in the mechanism of Sun 
Arcs and Rotary Spots, quietness of operation was obtained which 
overcame the objections of the sound technicians. The old type of 
broadside lighting units, used extensively in the days of silent pic- 
tures, has been practically abandoned in modern picture production, 
because its design inherently prevented silent operation. 

Early this spring, one of the leading producers of colored motion 
pictures requested Mole-Richardson, Inc., to investigate the possi- 
bility of developing a motion picture arc lamp for use as a general 
lighting unit. This firm was developing a new process of color 
photography, and it seemed that arc illumination would provide the 
most satisfactory means of lighting the sets to be photographed. 

* Presented at the Fall, 1933, Meeting at Chicago, 111. 
** Mole-Richardson, Inc., Hollywood, Calif. 


52 P. MOLE [j. S. M. P. E. 

The specifications were as follows : 

(1) The lamp should produce an illumination level of 200 foot-candles, as 
measured at fifteen feet with a standard Weston photometer. 

(2) It must have a comparatively flat distribution curve over a projection 
angle of sixty degrees or more, and the field of illumination should be devoid of 
any hot spots, i. e., areas of illumination that are photographically objectionable. 

(3) The feeding mechanism of the lamp should be so designed as to provide a 
reasonably uniform level of light intensity during its period of operation, and the 
spectrum of the light emitted should not show any alteration of its characteristics 
during the period of operation. 

(4) It should be silent in operation, so that it may be satisfactorily operated 
in conjunction with modern sound recording apparatus. 

(5) It should take such a form, and be so mounted, that it will be convenient 
for placement, and be of such weight as to be easily handled on the set. 

(6) It should be economical in operation with regard to attendance, the con- 
sumption of current, and carbon electrodes. 

First experiments were made with half-inch white flame carbons, 
which had previously been used in practically all arc broadside 
lighting units. By improving the reflecting surfaces, it was found 
possible to boost the light flux of the old type of broadside units from 
60 foot-candles, measured at fifteen feet with a standard Weston 
photometer, to about 90 foot-candles. 

We quickly realized, however, that even though we overhauled and 
installed new reflectors in our old side arcs, it would be impracticable 
to attain the 200 foot-candle requirement desired by our client. 

We communicated with the National Carbon Company to ascertain 
what new developments had been brought forth in arc carbons, which 
would be suitable for use in equipment of the broadside type, and 
as a result obtained samples of several types of carbons that were 
thought to fulfill the requirements. After numerous experiments, we 
decided that probably the 8-mm., special, copper-coated, cored carbons 
that were recommended would best suit the purpose. 

An old type broadside unit was adapted to operate with the carbons 
and was supplied with chromium-plated metal reflectors. With the 
8-mm. carbons in both the upper and the lower carbon holders a 
marked improvement in light intensity was attained. Utilizing prac- 
tically the same current, 40 amperes, the light intensity was raised 
from 90 foot-candles, measured at fifteen feet, to 120 foot-candles. 

An inherent fault of the old type of broadside arc lighting unit was 
its inability to maintain a uniform level of illumination. When first 
energized, the old type lamps would consume from 40 to 45 amperes, 


and produced their maximum lighting intensity; by the time the 
feed mechanism came into operation, the current in most cases 
dropped to approximately 32 amperes, and the lighting level 
dropped about 40 per cent. The specifications set forth demanded a 
much more accurate control of the lighting intensity. 

Knowing the limitations of the old style carbon control mechanism 
of the various lamps that had been previously designed, it was de- 
cided to experiment with a lamp in which each pair of carbon elec- 
trodes would be separately controlled. An experimental model was 
built, and after a number of modifications a mechanism was developed 
that reduced the fluctuations in light intensity during the feeding cycle 
of the lamp to within 10 per cent. 

In previously designed broadside lamps it had been attempted to 
control the feeding of the carbon by means of a single current coil in 
series with the arcs, and by utilizing various means for equalizing 
the feeding of the upper carbons toward the lower carbons. As far as 
we have been able to observe, mechanisms operated on such a prin- 
ciple fail to provide good operating conditions, due to the fact that 
the tolerances in the diameters of the carbon electrodes must of neces- 
sity be rather large; and if it happens that a carbon with a minus 
tolerance be placed in one side of the twin arc mechanism, and a car- 
bon with a plus tolerance be placed in the other side, the carbon hav- 
ing the small diameter will inevitably feed more rapidly than that hav- 
ing a larger diameter. It is most difficult to devise a mechanism oper- 
ating with a single control coil that would overcome the difficulty 
without greatly complicating the structural characteristics of the 
feeding device. 

The mechanism developed for the M-R Type 29 twin arc broad- 
side controls each pair of carbon electrodes, independently maintain- 
ing the voltage drop across each pair of electrodes at 35 to 40 volts, 
and the feed of each pair of electrodes is independent of the other 
and controlled by the voltage drop in the arc that the mechanism 

Fig. 1 is a schematic diagram showing the method by which this 
is accomplished. Each carbon arc has its lower carbon electrode in a 
fixed position. The upper carbon electrode is movable; and when no 
current flows, the lamp is in contact with the lower carbon. When 
the lamp is connected to the line, the circuit is closed with only the 
ballast resistance to impede the flow of current. 

The current coils of each mechanism are in series with each other 



[J. S. M. P. E. 

and with the two arcs. The current from the positive side of the 
line passes through the ballast resistance, 1, into the base of the lamp, 
through the switch to the control coil of mechanism No. 1, and on to 
the upper carbon; thence to the lower carbon, into the current 
coil of mechanism No. 2 through the coil to the other upper carbon, 
then to the lower carbon, and back to the line through the ballast 
resistance, 2. The energizing of the circuit actuates the solenoid 
armatures, which, through their connecting linkages, elevate the 
upper carbons in each arc system, striking both arcs. 


FIG. 1. Schematic diagram of arc regulating 

Above each current coil, and surrounding each armature, is a coil 
wound with fine wire and a large number of turns, connected across 
the arc controlled by it. These coils are wound counter to their re- 
spective current coils, and the instant the arc is struck a small cur- 
rent flows through each coil. Since they are shunted across the arcs, 
the energy introduced into them increases as the voltage drop of each 
arc increases, the magnetic flux of each voltage coil opposing that of 
its corresponding current coil. By properly proportioning the num- 
ber of turns in the current and voltage coils, and proportioning and 
spacing their respective armatures, it is possible by this method to 


control the opening of the arc and to maintain quite accurately a uni- 
form voltage* drop across the arcs. Ball-bearings were introduced 
at the fulcrum of the upper carbon actuating levers, so as to make the 
mechanism sensitive to the changes of voltage of the arc. Simple, 
plate-type carbon clutches have proved entirely adequate. 

Since maximum efficiency with the carbon electrodes used was at- 
tained by using a 5 / 8 -inch arc gap, it was necessary to take precautions 
to prevent magnetic "blowing" of the arcs. This was accomplished 
by connecting the current coils of each mechanism so that they formed 

FIG. 2. Broadside Twin Arc Lamp 
M-R type 29. 

a closed magnetic circuit, and by placing a steel magnetic baffle plate 
between the coils and the arc. 

The entire mechanism is relatively simple, and may be economically 
manufactured, because, except for connections in the wiring, each unit 
of the mechanism is an exact duplicate of the other. To adjust each 
mechanism so that it will operate in harmony with its adjacent unit, 
it was desirable that a simple adjusting means be provided. This ad- 
justing means is the movable counterweight mounted on the arc 
actuating lever. As the lamps leave the factory they are adjusted for 



[J. S. M. P. E. 

operation on 115 volts, 40 amperes d-c., voltage readings being taken 
across each arc and the counterweights adjusted for balanced opera- 

Under practical and test conditions it has been found that with this 
mechanism flickering has been totally eliminated. Even though the 
line voltage be greatly disturbed, as it often is on motion picture stages 
when operating under heavy loads, the mechanism is so responsive 

FIG. 3. Twin Arc Scoop Lamp M-R type 27. 

that such disturbances are compensated without the "blinking" that 
was often experienced with the old type twin arc lamps. 

The mechanism has been built into two types of lamp heads : the 
M-R type 29 Twin Broadside Arc and the M-R type 27 Twin Arc 
Scoop. The Broadside Lamp, designed for floor use, is mounted on 
a pedestal having two telescoping sections, and may be elevated from 
a height of four feet one inch, to eight feet eight inches from the floor. 
The housing of the M-R type 29 has been constructed of duralumin 
sheet metal and aluminum castings (Fig. 2). The mechanism may 
be tilted from the vertical position thirty degrees forward or backward 
without disturbing the operating characteristics. Chromium plated 


reflectors, which have proved to be entirely satisfactory in this type of 
equipment, increase the light flux of the lamp in excess of the speci- 
fication requirements. 

The scoop is illustrated in Fig. 3. Its housing, in addition to carry- 
ing the mechanism, also carries the resistance units. To facilitate the 
dissipation of the added heat of the resistance, the head has been 
amply ventilated with louvers. The aperture of the lamp has been set 
at an angle to deflect the light downward, as the scoop is primarily 
designed for overhead use. To assist in carrying off the fumes from 
the arc coring, both types of lamps are provided with a chimney 
midway between the twin arcs. This ventilation contributes to the 
cleanliness of operation of the equipment, a large portion of the white 
condensate from the arcs passing off through the chimney. 

Both types of lamps are intended to be used with glass diffusers. 
A prismatic glass, sand-blasted on one side, has proved best for the 
purpose, its high lead content inhibiting the transmission of ultra- 
violet radiation. No complaints have been received from actors 
working under the lamps in regard to injury of their eyes. 

While it is not anticipated that this new equipment will revo- 
lutionize motion picture stage lighting, there are many types of 
photography and many special effects for which this equipment is 
peculiarly adapted. 


MR. JOY: Any one who has seen Mr. Mcle's lamp in operation realizes that 
he has made a very material contribution to the art of illuminating motion pic- 
ture sets. The feeding of the carbons is uniform and regular, as the feeding sole- 
noid of each arc is controlled by the current and voltage of that arc, resulting in a 
steady light from the unit. Tests have shown that within an angle of 60 degrees 
in front of the lamp, the decrease in light from the center to the outside is only 
about 15 per cent. Such a small change over such a wide angle should be par- 
ticularly advantageous in photographic and motion picture work. 

MEMBER : What is the bulk or weight of the equipment? To what extent does 
0. add to or detract from the regular incandescent equipment? 

MR. MOLE : It would not add to the bulk or the number of units. Experience 
has shown that the number of units used on the set depends entirely on the set, 
regardless of whether arcs or incan descents are used. As many units are used as 
the size of the set demands, so that the entire set will be covered. 

MR. COUR: What is the comparison in wattage? 

MR. MOLE: That is very difficult to answer. One cameraman on a 15 by 15 
set would use 600 amperes and another would use 1200 amperes, so there is no 
way of determining the saving. More lumens per watt are radiated by an arc 
than by incandescents, but whether a man is working on a low level or high level, 
we don't know. 




Summary A new type of white flame arc for use in photography is described 
and its distribution of energy and the composite effect of this energy distribution 
on the transmission of the lenses and the sensitivity of super sensitive panchromatic 
film are shown. Of the total radiant energy almost 37 per cent is in the visible portion 
of the spectrum, and the new carbon shows a very considerable increase in light 
emitted over former carbons at the same current and voltage. 

Carbon arcs were the first artificial light sources used for photog- 
raphy of various kinds where light was required for more than a very 
few seconds. The first arcs used were d-c. plain carbon arcs in which 
the source of light was the incandescent crater of the positive carbon. 
These plain carbon arcs were relatively inefficient from the standpoint 
of photographic power per watt of electrical energy, since only about 
17 per cent of the radiant energy was in the visible spectrum and 
the photographically effective ultra-violet. 

About 1910, the white flame carbon arc was introduced to the 
photographic industries largely as a result of the work of one of the 
early members of this Society, William Roy Mott, of this laboratory. 
The source of light in this arc is the brilliant flame between the 
electrodes, which themselves give very little light. The visible and 
photographically effective light from this arc is from 30 to 35 per cent 
of its total radiant energy. In addition to white, the flame arc can 
be made to produce light of other colors which have been useful for 
certain types of photographic work. 

The effectiveness of a light source in photography is dependent 
upon the distribution of its radiant energy throughout the spectrum, 
the spectral sensitivity of the photographic emulsions, and the trans- 
mission factors of the lenses used in the camera. 1 

In a previous paper 2 the spectral energy distributions of the 
regular white flame photographic carbons and the panchromatic O 

* Presented at the Fall, 1933, Meeting at Chicago, 111. 
** Research Laboratories, National Carbon Company, Inc., Cleveland, Ohio. 



carbons at 35 amperes and 37*/2 volts on direct current with the 
upper carbons positive were presented, together with the efficiencies 
in lumens per watt of those and several other flame arcs. There 
were also presented the photographic effects of the use of these car- 
bons upon the panchromatic film in use at the time. Subsequently, 
the supersensitive panchromatic film was introduced to the motion 
picture industry and the effects of various light sources upon the proper 
rendition of colors were given in an article by Finn, 3 in which it was 
shown that the white flame arc with this supersensitive panchro- 
matic film gave better definition and better rendition of colors than 
other light sources. 

An improved white flame photographic carbon has been developed 
in the laboratories of the National Carbon Company known as the 













FIG. 1. Spectral energy distribution of 8-mm., copper-coated, 
motion picture studio carbons: 35 and 40 amperes, 37.5 volts, d.-c.; 
each rectangle represents 10 microwatts per sq. cm. at one -meter 

"National copper-coated M.P. studio carbon." This new carbon 
is 8 millimeters in diameter and 12 inches long, with a core of suitable 
size for flaming arcs filled with a composition of rare earth chemicals 
of the cerium group. The cross-sectional area of this new carbon 
is only about 38 per cent of that of the well known 13-mm. X 12- 
inch white flame photographic carbons. Since the small carbons 
carry the same or greater current, it is necessary to coat them with 
copper. This, however, does not impair the steadiness of the arc, 
which is always better with small carbons when compared with 
larger sizes at the same current. Fig. 1 gives the spectral energy 
distributions at 37 1 /* volts and 35 and 40 amperes on direct current 
with the upper carbon positive. 

The comparison of these spectral energy distribution curves with 



[J. S. M. P. E. 

that for the 13-mm. white flame photographic carbons given in Fig. 1 
of the previous paper 2 shows that while the radiant energy in the near 
ultra-violet is essentially the same, the radiant energy of the new 
carbon in the visual portion of the spectrum is at a considerably 
higher level and shows a peak in the region where the supersensitive 
panchromatic film is least sensitive, which should make for desirable 
photographic results. 
















- - 



70 fc 

1 \ 





































><xr " 4600 50 

Wavelength Angstrom Units. 

FIG. 2. Photographic effect vs. wavelength, for supersensitive 
panchromatic film: broken curve, sunlight; solid curve, 8-mm., 
copper-coated motion picture studio carbons: 40 amperes, 37.5 
volts; ordinates represent the product of transmission of glass 
lens, relative sensitivity of film, and spectral distribution of light 
source values, adjusted to maximum of 100. 

Table I shows the percentage distribution of radiant energy from 
these carbons at 40 amperes and 37 1 /z volts, with the upper carbon 


8-Mm., National Copper-Coated, M.P. Studio Carbons 
Relative Energy in Several Spectral Bands 

Visible light, 4000-7000 A. 

Photographically effective, 3400-7000 A. 

Infra-red (heat), 7000-50,000 A. 

Total radiant energy of wavelengths less than 50,000 A. 

Per Cent 





From this table it is seen that only 58.4 per cent of the radiant 
energy from this new M.P. studio flame arc is heat. These (8-mm., 
copper-coated, M.P. studio) carbons at 40 amperes and 37.5 volts 
give about 9330 candle-power in the horizontal direction, and are 
consumed at an average rate of 3.6 inches per hour. 


A twin arc (two arcs in series) without reflectors will give approxi- 
mately 200,000 total lumens at the same current and voltage com- 
pared with 158,000 total lumens for the older 13-mm. white flame 
carbons under the same conditions. Such a twin arc burning the 
new 8-mm. carbons gives an efficiency of approximately 40-46 
lumens per watt on a 11 5- volt line as compared with 34-38 lumens 
per watt with the 13-mm. carbons in the older studio twin arc lamps 
without reflectors. 

Fig. 2 shows the photographic effect in relation to wavelength for 
supersensitive panchromatic film with sunlight and the copper- 
coated M.P studio carbons at 40 amperes and 37 1 /2 volts. 

These curves of photographic effect take into account the spectral 
sensitivity of the supersensitive panchromatic film, the transmission 
of the glass camera lenses, and the spectral energy distribution of 
the light source. The calculations were made by the method de- 
scribed by Jones. l 

Fig. 2 shows very clearly that the photographic effect of the light 
from the new studio carbon arc follows the form of the photographic 
effect of sunlight very closely indeed. We believe that the energy 
distribution curve and the curve of photographic effect of the radiant 
energy from these studio carbons show that they should be a very 
desirable source of illumination for all kinds of photography. 


1 JONES, L. A. : "The Use of Artificial Illuminants in Motion Picture Studios," 
Trans. Soc. Mot. Pict. Eng., V (1921), No. 13, p. 74. 

2 JOY, D. B., AND DOWNES, A. C.: "Characteristics of Flame Arcs for Studio 
Lighting," Trans. Soc. Mot. Pict. Eng., XII (1928), No. 34, p. 502. 

3 FINN, J. J. : "Soft and Hard Lighting with Supersensitive Panchromatic 
Film," Internal. Proj., 2 (March, 1932), No. 1, p. 33. 


H. B. LEMON** 


Summary. Committed to a policy of undergraduate education of a general type 
during the first two college years, the College was faced with the problem of presenting 
the natural sciences to the entire undergraduate body by means of lectures only. A b- 
sence of effective laboratory work seemed to doom the enterprise to failure. Demon- 
stration lectures have serious limitations with large groups. The talking motion 
picture is a perfect medium to use in support of the demonstration lecture in these 
two ways: close-ups of delicate apparatus can be projected on gigantic scale; natural 
large-scale phenomena out of doors, and industrial processes may also be brought 
vividly into the classroom by this means. 

The University of Chicago is engaged in producing a series of films designed 
specifically for the four Introductory General Courses of the University, viz., the 
humanities, and the social, biological, and physical sciences. Completed and already 
used with excellent success are films on the molecular theory of matter, oxidation and 
reduction, energy and its transformations, and electrostatics. In production to be 
ready for use this year are two reels on wave motion and sound, magnetism and elec- 
tromagnetism, one reel each. Films on the velocity of chemical reaction, chemical 
equilibrium, atomic and molecular structure, spectra, interference of light, the velocity 
of light, carbon and its compounds, the atmosphere, the solar system, the changing sur- 
face of the earth, weather and forecasting, time and the calendar, and volcanic phz- 
nomena and earthquakes are among those planned for in the physical sciences. 

This changing world of politics, industry, and economics that we 
have been going through has had its counterpart in a changing world 
of education. In the year 1892 about 200,000 students were enrolled 
in the high schools of this country. Forty years later, in 1932, 
over 4,000,000 students were enrolled in the high schools of this 
country. The percentage of our population of high-school age who 
went to high school was 10 per cent thirty years ago, and is now over 
50 per cent. While these figures relate to high schools, the situation 
in the colleges, while not involving nearly such large numbers of 
students, has been closely parallel, and those of us who have been in 

* Presented at the Fall, 1933, Meeting at Chicago, 111. 
** Professor of Physics, University of Chicago. 


touch with undergraduate and graduate education in our universities 
in the last ten or fifteen years, have been aware of serious problems. 

The business of going to college had become, up to two or three 
years ago, very largely a social matter, whereas going to college in the 
days of our parents was solely a matter of acquiring a very special 
training preparatory to entering one of the learned professions. 
In recent times, going to college has seemed to be something in the 
nature of having four years more of fun before having to go to work, 
and many of us have felt in all seriousness that a very large number of 
our college men and women were learning, in the four years of college, 
chiefly bad habits. I don't mean by that anything involving actual 
deeds of moral turpitude, but very bad habits in preparation for life 
such bad habits, as never to do anything thoroughly well, or de- 
veloping a chief interest in social activities. While I would never 
advise any young person to let his studies totally interfere with his 
college education, it seemed to us that there was not at all the proper 
motivation on the part of many of these young people. Of course, 
we never blamed the young people for that; we blamed the parents, 
we blamed ourselves. It may be that one of the silver linings of this 
very dark cloud under which we were passing is going to be a very 
marked change of attitude in the future toward this business of going 
to college. 

As a result of that and other things, the University of Chicago two 
years ago adopted a new plan of college education, a plan that very 
frankly avoided the training of specialists in the first two years of 
college. We are not attempting to train specialists in physics or 
chemistry or mathematics, hardly even starting that training at the 
University of Chicago in the first two years. We are devoting the 
first two years to what we call general education. There is a danger 
in such a procedure. There has been some tendency in recent years 
to teach students more and more about less and less. General 
education survey courses, as they are sometimes called have a 
tendency to do just that sort of thing. We wanted our curriculum 
to have depth, as well as the utmost breadth. 

Consequently, in the first two years at the University, every stu- 
dent is required to take, out of a total of eight courses in the two 
years, four general courses one in the physical sciences, one in the 
biological sciences, one in the humanities, and one in social studies 
each of which require a full year. That makes four of the eight. 
In addition, two full-year, second-year courses in two of those 

64 H. B. LEMON [J. S. M. P. E. 

four fields are required. So our undergraduate body in the first 
two years will have covered by intensive study during two years, 
one-half the entire field of university activities, and all of it in one- 
year courses. That takes six out of the eight courses to do that and 
it leaves them two courses available for continuation of interests 
in high-school English, special courses, modern languages that 
they may need. 

These changes, of course, involved an entirely different problem 
for our teaching force, an entirely new curriculum, and very largely 
an entirely new staff, a staff which now, instead of being largely in 
the hands, as in the old curriculum, of relatively young and inexperi- 
enced men, required the drafting of a personnel that had more 
experience, and in so far as possible represented distinguished work, 
both in research and in teaching. 

Coming now to the very definite problem of the physical sciences 
we were offering a new course in addition to all the old courses we 
already had, in which the numbers of students involved is enor- 
mously greater than ever before. Students going to the University 
under the old curriculum interested in social sciences and humanities, 
students in general, avoided physics for the very good reason that 
our whole courses in physics were designed for the training of spe- 
cialists. Now they all have to take not only physics, but chemistry, 
astronomy, geology, and mathematics, in this survey course. 

One of the problems to be faced was the question of how to present 
our source material in the physical sciences. One would not know 
how, I believe, to teach students something about the interpretation 
of literature, or how to project a study of literature before a group 
of students who had never read any books. As books are the source 
material for literature, so are phenomena the source material for 
science. The conventional laboratory was out of the question. 
The freshman and sophomore students a vast majority of them 
had never been inside a laboratory. Even if they had, the problem of 
administering elementary laboratory work to a class of 750 students 
was impossible. It is in the laboratory, in conventional courses in 
physics and chemistry, that students acquire most of their first- 
hand knowledge of the phenomena. 

We knew perfectly well from former experience that we could not 
contemplate giving a "talkie on science" getting up and talking 
about science. The words go into one ear and out the other, and 
leave no impression. We already had the technic of demonstration 


lectures where you demonstrate the phenomena on the lecture table. 
Statistics show that in the last thirty years the use of the demon- 
stration lecture has greatly increased. 

Experiments on the lecture table would probably be clearly visible 
to an audience of this size, under the best conditions, about half-way 
back in the room. Those at the rear could not see ordinary appara- 
tus, so at once we were limited to showing on the lecture table 
things that could be shown with fairly large-scale apparatus; and, 
of course, physical laboratories whose main business has been re- 
search all these years, have no such equipment. There are notable 
laboratories in Europe where demonstration lectures can be made to 
audiences as large as 400 or 500 persons, and the audience can see 
everything that is going on because the apparatus is all built on a 
large scale, very brilliantly illuminated. This, obviously, was one 
way of presenting source material. 

Another method is to use the permanently set up demonstrations in 
very elaborate museums of science. The sort of thing you have 
been seeing in the Hall of Science at the World's Fair can be done on 
a smaller and less ambitious scale locally, and devices operated by the 
students, on which they press a button and an experiment takes place, 
resets itself, and is repeated as often as the button is pressed, are 
very effective means of presenting phenomena of science. 

We have such museums one very elaborate one in physics, one 
less ambitious which we hope to develop further for chemistry. The 
Adler Planetarium serves as our astronomy museum, where our classes 
meet at certain stated periods. We have a museum of geology in 
which are shown many of the things seen in the geology exhibits at 
the Fair. Indeed, not a few of those exhibits have taken form 
from what we have been doing at the University, and have been re- 
built for the Hall of Science. But there are many things that can not 
be done by any of those methods ; but, as a matter of fact, the demon- 
stration lecture table can be greatly assisted by the talking motion 
picture. We are not speaking academically or from theory. We 
know it can because we have done it. 

There were no films available that were suitable for our courses. 
Not being professional motion picture producers, we were not compe- 
tent to make such films. We knew what we wanted, but how to get 
it was another matter. Electrical Research Products, Inc., having 
already studied the technic of educational motion pictures, assisted 
in the production. We, the faculty, familiar with the university 

66 H. B. LEMON [J. S. M. P. E. 

activities and the nature of the general courses, had already worked 
out a comprehensive plan from the point of view of the desirable 
content of the pictures. We have tried only to produce pictures 
that were exactly what we wanted to use in our classes. The films 
that were produced may find a wide usefulness in other state and 
privately endowed universities. They will be equally useful in small 
colleges and ultimately, if our plans go ahead and the reels expand 
somewhat, they will have an enormous usefulness in the secondary 
schools. We have already used four films in our courses and have 
four more in production that we expect to finish in time for use this 

We have only one set of data, one year's class; but from those data 
we are firmly convinced that we can cover more ground in less time 
with a class, and make the information stick very much better. 
That, of course, is the ultimate aim of the teacher, always. As the 
knowledge increases we are continually faced with the need for cover- 
ing more ground and, of course, we always want to make it stick 

Let me tell you something of the manner in which we use our films 
in the class. Imagine that you were in a class, and that you had 
never heard anything about the molecular theory of matter. You 
would first be presented with a synopsis of the entire subject covered 
in the film. At the first meeting, in the first ten minutes, you would 
watch the film and listen to it. The lecture would be fifty minutes 
long. Forty minutes would remain in which we should arrange 
supporting apparatus for additional demonstration on the table; 
so that after you had seen the film, we would show you some experi- 
ments that would not be duplicates of those in the film, but things 
that could be done effectively on the table. On the next day we 
would talk to you for forty or fifty minutes again. Day after that 
the subject would still be on the boards, and we would probably 
begin the performance by running the film silently and talking to 
you about it, pointing out many things that had escaped your 
attention at first. The rest of the hour would be consumed by talk 
on our part, questions on yours. When the subject had been fairly 
well concluded, in the last ten minutes of the last hour you would 
see the talking picture once more, with the sound audible, as a review. 

We give these details partly because the general feeling on the 
part of teachers is that when the teacher advocates talking motion 
pictures for use in classes, he is organizing himself out of a job. 


These films that we are making are utterly useless without a teacher. 
They constitute nothing more in our curriculum than an additional 
and a very effective tool, which, together with the museum, together 
with the demonstration lecture, together with the quizzes, the recita- 
tions, and all the old tools that we had, enable us to do a better and 
much more comprehensive job, in very much less time. 


MR. FRITTS: What is the relative value of 16-mm. vs. 35-mm? Also, as- 
suming that in due time we shall have, in addition to black-and-white, natural 
color and three dimensions, of what value would they be? 

MR. LEMON: With respect to your first query, we use at the University 16- 
mm. sound-on-disk, Western Electric equipment, in my own lecture room. We 
have available for use, also, for larger groups, 35-mm. sound-on-film. So far 
as my two years of experience goes, I see no choice. The 16-mm. film is probably 
not capable of projection on quite as large a scale; but for audiences the size of 
our classes 100 to 250 persons the 16-mm. equipment is perfectly effective 
in our own lecture rooms, which is acoustically treated. 

With respect to color, my reaction would be, speaking academically and in 
advance of experience, that the more realistic the picture on the screen can be 
made, the better it will be. However, we felt, before trying these pictures, that 
photographed experiments would not go across to a class as well as actual experi- 
ments on the table. However, when the film is used in conjunction with demon- 
stration experiments, we find that the students are apparently quite unconscious 
of the difference. The black-and-white seems to go across quite as well as the 
actual experiment on the table. 

I want to emphasize that we have as yet only one set of data, on one year's 
course. Any conclusions that I may draw now, since this is frankly an experi- 
ment in education, are subject to revision as data accrue. I see no present need 
for color; there is much to be done before that, in any event. 

MR. FRITTS: At one time it was suggested that a teaching projector should 
be capable of being stopped at any particular point for further elucidation by the 

MR. LEMON: That is very desirable; when we show silent film we do stop 
it occasionally. However, it is rather diastrous to the film, for the heat filter 
is not quite adequate. We can't stop too long, so I always have my assistant 
at the machine to stop it long enough for me to point out everything I want to 
point out. 

It is very difficult to talk to one of these pictures. No one who hasn't actually 
done it can appreciate the enormous amount of time and effort that go into the 
selection of every particular word, and the problem of synchronization. To 
have the correct phase of the picture meet the eye just as the corresponding con- 
tent strikes the ear that is when the context is appreciated is an extremely 
difficult thing. 

For the sake of some schools, and because of the present financial situation 

68 H. B. LEMON [j. s. M. P. E. 

of most educational institutions, we have to cover the temporary emergency 
with a study guide, which contains a certain amount of instruction about use, 
and so on, and suggestions for the teacher, but the most important part of which is 
the full-scored script of the film. Consequently, a school without sound equip- 
ment at the present time, wanting to use these films, will find it relatively easy 
to have the instructor simply read from the study guide while the film is being 
run silently, and he can synchronize fairly well. 

MR. RICHARDSON: Do I understand that the University of Chicago has been 
convinced of the fact that in the future the school or college that attains the 
highest proficiency in education must have special talking films for certain 

MR. LEMON: One hesitates to make predictions. The basic assumption 
that we are entertaining in these new general courses is that in the future prob- 
ably not nearly as many students will go to college for four years. We are rather 
expecting that large numbers of our students will leave the University, at least 
for some period of years, at the end of their sophomore year. They will then 
have been trained in no specialty; they will have received no technical training 
at all, but will have been exposed for two years to the entire scope of University 
activities. However, they will be in a much better position than many of the 
four-year graduates now are, who find that the classical curriculum has not given 
them even a background for solving the problems they have to face in life. 

We expose our students for two years to the whole scope of our activities, 
and for an additional year to half the scope. In order that a respectable number 
of our students do creditable work on comprehensive examinations on these 
subjects, at the end of a year's time, we have resorted to the talking motion 
picture as an additional tool in our educational problem. Otherwise, we should 
never have discovered how effective the pictures can be. 

MR. MATTHEWS: As I viewed the two demonstration films, it seemed as though 
the tempo of the second film was more suitable, as far as receptivity was con- 
cerned, than that of the first. I thought that the speed of the former was too 
great: one idea followed another so rapidly that in listening it was difficult to 
absorb the idea that was intended to be conveyed before another idea was pre- 

MR. GREENE : One of the great advantages of film methods of presentation 
is that the viewing distance can be shortened to any extent desired. The in- 
structor can, in effect, bring his class right down beside him, instead of having 
them view the operations from a remote point at the rear of the lecture hall. 

As to the matter of three-dimensional pictures, the difficulty attending the 
use of supplementary apparatus can easily be obviated in the classroom, whereas 
it is very detrimental in the theaters. The instructor can easily instruct his 
class on how to handle the auxiliary apparatus, and since his audience consists 
always of the same persons, at least during a given term, there should be no 
difficulty in doing so. 

As to the matter of holding the film stationary at any point, would it not be 
practicable, where the pictures are repeated term after term, to provide lantern 
slides for those frames? If not, a light source of higher temperature might be 
used in the projector; if the installations are permanent, a high- temperature arc 


might be used instead of a Mazda lamp. It would be a bit more difficult to 
control, but would furnish a light that is much cooler. 

MR. HOLSLAG: I believe that the use of color would benefit such a film con- 
siderably, because color could be used, in animation, to differentiate between two 
important curves or points, such as is sometimes done in the case of graphs. 
The color and sound would be supplementary, and would convey the ideas more 
forcibly than they could be conveyed by color or sound alone. 

Speaking of stopping the film, or slowing it down, it seems to me that an un- 
fortunate psychological effect would be produced upon the class: we are all 
familiar with the queer sound that is produced when the sound track is slowed 
down. A much better method would be simply to duplicate the desired frame 
a number of times on a special printer. This would produce the same effect as a 
slide without the attendant difficulties of stopping the film and of bringing an- 
other projector into action. The instructor could determine beforehand what 
particular frame to duplicate, and it might be made to cover three to five feet, 
or whatever length might be necessary. 




The use of motion pictures in the non-theatrical field, such as in schools 
colleges, churches, commercial organizations, steamship lines, hotels, hospitals, 
etc., was well established for many years before synchronized sound found its 
place in the motion picture industry. This field was well supplied with various 
types of portable motion picture projectors, among which was one known as the 

Projectors for silent film, however, are no longer saleable; for that reason the 
International Projector Corp. has developed a portable 35-mm., sound-on-film 
equipment that admirably takes the place of its predecessor. The object of 
this paper is to describe briefly the design and construction of this new equipment. 
It is assembled completely in two carrying cases (Fig. 1), in one of which are the 
projector and sound reproducing mechanism. The amplifier, loud speaker and 
cable, and the upper magazine of the projector are carried in the second case. 
The front of the case is used as a baffle for the loud speaker. 

Considerable effort has been expended in order to combine high-quality equip- 
ment with its correspondingly accurate assemblies, with acceptable portability. 
The entire projector mechanism, lamp house, take-up magazine, motor, and sound 
head are enclosed in a substantially built carrying case 22*/2 inches long, 24 inches 
high, and IQ 1 /-* inches wide, and the equipment is so constructed that no parts or 
adjusting mechanisms project beyond the confines of the case. This feature 
eliminates the possibility of damaging the apparatus during shipment or when 
carried from place to place. 

The projector is of the straight feed type, similar to standard professional equip- 
ment, and the case door is provided with two glass observation ports in order 
that the film may be observed while in transit through the equipment. The door 
is provided with a lock and key to prevent unauthorized persons from tampering 
with the equipment. The film magazines satisfactorily accommodate 1000 feet 
of standard 35-mm. film, and are constructed from one piece of metal with no 
soldered joints. 

The conventional fire valves are provided on both the upper and lower maga- 
zines, and the magazine doors are substantially supported by heavy hinges. A 
spring latch is provided to hold the magazine doors closed. The upper magazine 
is attached to the top of the mechanism case by means of two thumb screws, and 

* Presented at the Fall, 1933, Meeting at Chicago, 111. 
** International Projector Corp., New York, N. Y. 



is readily removed for packing in the amplifier case. Professional practice is 
followed in the construction of the upper magazine ; it is equipped with a revolving 
spindle having a key and toggle latch for holding the reel, and an adequate 
hold-back tension device to prevent the film from over-running during projection. 

Provision has been made for holding both the lower magazine door and the 
projector case door open while threading, regardless of the projection angle of the 
projector; and the take-up magazine door is positively closed and latched with 
the closing of the door of the projector case. 

The projector (Fig. 2), sound mechanism, and take-up unit are entirely gear 
driven. No belts or chains of any kind are used in any part of the equipment. 
Bearings of all drive-shafts other than the intermittent movement are of a bronze, 
oil-absorbing composition, and oil fed to the outer surface of these bearings seeps 
through the bearings to the spindles, thus providing adequate lubrication at all 

FIG. 1. Right: case containing projector and sound mechanism; left: 
case containing amplifier, loud speaker and cable, and upper magazine of 
the projector. 

times. Oil is fed to these lubrication channels by means of conspicuously located 
oil tubes and oil holes in the several parts of the apparatus, and these lubrication 
points are shown in a threading and oiling chart permanently attached to the 
door of the projector. 

The revolving cut-off shutter is placed between the lamp house and the projec- 
tion aperture and is a little larger than 8 inches in diameter. The shutter is en- 
closed by a guard so as to protect the projectionist from injury. Provision is 
made for setting the shutter without removing any part of the equipment other 
than the guard. This construction offers the same advantage and protection, 
in so far as heat reduction at the aperture is concerned, as is gained by rear shutter 
equipment on professional apparatus. 

The fire shutter is of the centrifugally operated type, which opens automatically 
when the projector has reached a safe operating speed, and closes automatically 



[J. S. M. P. E. 

when the projector is stopped. No friction operated devices are used in this 
connection, and the unit is to all intents and purposes the same as is used on the 
Simplex projector. 

An automatic safety trip is provided for the fire shutter (Fig. 3). Should the 
film for any reason stop at the aperture plate while the projector is in operation, 
the fire shutter is instantly released, thus preventing the film from taking fire 
at that point. This safety trip is of the automatic resetting type so that all its 
component parts are automatically restored to their operating positions when 
the projector comes to rest and the reason for the stoppage has been removed. 

FIG. 2. View showing gearing of projector sound mechanism 
and take-up unit, filter flywheel, and switches. 

The intermittent movement is of the conventional Geneva type, except that 
the star-wheel, cam, and the cam pin are hardened and ground on all operating 
surfaces, and the assembly operates in a fluid oil bath. The housing for the inter- 
mittent assembly contains also the driving units for the upper feed sprocket 
and the shutter shaft, all of which are lubricated by the splash system of the 
intermittent movement assembly. An observation port is provided in the casing 
so that the oil level may be easily determined at all times. 

The picture is framed from outside the projector case by moving the entire 
intermittent mechanism laterally with respect to the aperture plate, allowing for 
the framing of one full picture, A framing and threading lamp is provided in such 

Jan., 1934] 



a position that the picture may be properly framed at the aperture when threading 
the projector. This lamp gives sufficient illumination for threading the projector 
in a darkened auditorium or projection room, and it is possible to turn the pro- 
jector over manually in order to position the intermittent movement properly 
for threading. The projected picture is exceptionally steady, due to the special 
construction of the film trap and gate assembly, which are also designed to project 

FIG. 3. View showing assembly in case. The automatic 
safety trip is shown immediately to the left of the upper feed 
sprocket; to the left of the trip is the fire shutter guard 
mounted adjacent to the lamp house. 

buckled film with a minimum of optical disturbance. The lens is focused con- 
veniently from outside the projector case. 

The lamp house (Fig. 4) is of the double-walled type with a ventilating space 
between the walls, and a gear-driven fan is provided in the top of the lamp house, 
which directs the ventilating draft and at the same time prevents stray light 
from leaving the lamp house. The latter is provided with a standard pre-focus 
Mogul socket, which will accommodate any standard T-20 pre-focus lamp, and 



[J. S. M. P. E. 

an adequately supported spherical mirror. Socket and mirror are rigidly held 
in their correct optical positions by means of adjusting mechanisms that may 
be securely locked after the various units have been correctly aligned. 

Switches of ample carrying capacity are provided for the projection lamp, 
motor, and exciter lamp, respectively (Fig. 2). The projection lamp and exciter 
lamp switches are mechanically coupled in such a manner that the lamps may not 
be operated separately for threading and running-out purposes, so that the 

FIG. 4. View showing lamp house, framing and threading lamp, automatic 
safety trip, and exciter lamp system. 

switch assembly provides a satisfactory means for making change-overs when 
two projectors are used. The usual amplifier change-over is avoided by this 
means, the actual change-over of the sound reproducing equipment being effected 
through the exciter lamps. 

A separate input receptacle is provided for the projection lamp, and an additional 
input receptacle is provided for the motor and exciter lamp circuits. The exciter 
lamp is heated by alternating current through a step -down transformer mounted 
in the equipment. An output receptacle is also provided, connected in parallel 

Jan., 1934] 



to the motor and the exciter lamp receptacle, this receptacle being used to supply 
alternating current to the amplifier. 

The driving motor is of the 115-volt, 60-cycle, split-phase induction type, and 
is mechanically connected to the mechanism through a flexible coupling. The 
motor is resiliently mounted in such a manner that its mechanical vibrations 
are effectively prevented from reaching the driven mechanism. 

The sound reproducing system (Fig. 4) is operated entirely by alternating 
current. The exciter lamp operates on 10 volts a-c., drawing 7 l / 2 amperes. The 
sound optical system (Fig. 4), when properly adjusted, projects a scanning beam 

FIG. 5. Amplifier, loud speaker, cable, and upper magazine; the loud 
speaker operates through an opening in the front of the case (Fig. 1), the front 
of the case acting as baffle. 

0.008 mil high by 0.084 mil wide. Means are provided for properly focusing the 
objective with relation to the sound track on the film without affecting the 
angular adjustment of the unit. 

The sound aperture plate is of the curved type, without tension shoes or gate, 
the tension on the film being supplied by a sprocket under tension mounted on the 
intermittent movement assembly. By this means the film is maintained in the 
correct optical plane with respect to the scanning beam. Means are also pro- 
vided for the instantaneous lateral adjustment of the film at this point. This 
adjustment may be made with the projector in operation, and when the standard 
S. M. P. E. sound test film is used for the purpose, the correct positioning of the 
sound track becomes a very simple matter indeed. 



[J. S. M. P. E. 

A radical departure from conventional filtering methods has been made in 
this equipment (Fig. 2). The sound filter flywheel, instead of operating at 360 
rpm., is operated at 1725 rpm. through an accurately machined pair of gears, 
the gear and flywheel floating within certain limits on the projector drive shaft. 
No flywheel forms part of the sound sprocket shaft assembly proper. This unit 
is therefore very simple to manufacture and performs very creditably. Great 
accuracy, of course, is necessary in this assembly, and all manufacturing toler- 
ances in this connection have been reduced to an absolute minimum. 

The sound sprocket operates in a free loop of film except for the tension sprocket 
referred to above (Fig. 3), and a hold-back sprocket is provided between the 

FIG. 6. Speaker carrying case equipped for double-unit installation. 

sound sprocket and the take-up, effectively preventing any variation due to the 
transmission of take-up tension to the sound system. 

The photoelectric cell, of the conventional type, is coupled to the amplifier by 
means of a low-capacity cable with an unbroken continuous shield from the 
socket of the cell to the amplifier input receptacle. 

The amplifier is of class A construction, and operates on 115 volts, 50 to 60 
cycles (Fig. 5). It has a maximum undistorted output of 5 watts (the harmonic 
distortion not exceeding 5 per cent), and is capable of reproducing all frequencies 
up to 9000 cycles. It is equipped with a special input jack for phonograph, a 
receptacle for two-button microphone, a low-frequency cut-off switch, a tone 
control for eliminating needle-scratch in phonograph disk reproduction, a high- 


and low-voltage switch, a-c. on and off switch, volume control, input receptacles 
for two projectors, monitor speaker receptacle, and auditorium speaker receptacle ; 
and all tubes are enclosed in a substantial perforated grille. 

The auditorium speaker is of the electrodynamic type, and is mounted in the 
speaker carrying case, the front of which acts as a baffle. The speaker opening 
in the case is provided with a removable cover to prevent damage during ship- 
ment. The speaker is provided with a 50-ft. length of cable. 

The speaker carrying case (Fig. 6) for a double-unit installation is slightly 
larger than that used for a single unit, for the reason that space is provided for 
carrying two upper projector magazines. This construction makes possible a 
transportable unit for a double projector installation consisting of only three cases. 
In all other respects the double- and single-unit speaker cases are identical. An 
entirely metal, rigid, collapsible stand with telescoping legs is available for this 
equipment, which, when collapsed for shipment, is approximately 24 inches 
long, 10 inches wide, and 2 3 / 4 inches high. 

The equipment is manufactured in three types : one for operation on 60 cycles, 
115 volts a-c., the second for operation on 50-cycles, 115 volts a-c., and the third 
for operation on either alternating current from 40 to 60 cycles, 115 volts a-c., 
or 115 volts d-c. The motor in the latter is electrically governed, is designed to 
operate at a fixed speed of 90 feet per minute for sound film reproduction, and 
provision is made through control equipment for reducing the speed so as to 
allow satisfactory projection of old silent film productions at the correct projection 



A meeting of the Board of Governors is scheduled for January 19th at New 
York, at which time the results of the balloting on the amendments of the Con- 
stitution and By-Laws proposed at the Chicago Convention will be announced. 
Among various other items on the agenda are the nominations of the new officers, 
provided the amendments are approved, the preparation of the 1934 budget, the 
launching of a vigorous membership campaign, and the formulation of plans for 
the Spring, 1934, Convention, which the Board voted at its last meeting is to be 
held at Atlantic City, N. J. 


The first meeting of the 1933-34 season was held in Los Angeles at the Maryland 
Inn and was convened at 7:00 P.M. at dinner. The meeting was attended by 
fifty-one members and guests, all of whom displayed marked enthusiasm and con- 
tributed valuable suggestions to the proposed agenda for the ensuing year. 
Chairman Emery Huse called the meeting to order, and appointed Messrs. Silent 
and Handley to tally the sealed ballots of the annual election for Section officers. 
Following their count, Mr. Silent announced that Mr. Huse had been reelected 
Chairman of the Section, Mr. Harcus had been elected Manager for the ensuing 
two years, and Mr. Rackett had been reelected Secretary-Treasurer. These of- 
ficers were installed with appropriate declarations of policy. 

The Chairman then invited Mr. Mole to express some thoughts that he had 
regarding the activities of the Section. Mr. Mole outlined the functions that 
the Section had performed since its formation, and stated that he felt that the 
general situation in Hollywood warranted greater activity on the part of the 
group, with particular reference to the important link occupied by the Society 
between the research laboratories of the equipment manufacturers and the studio 
technicians, who should be kept in close touch with new developments and recom- 
mended features which will make them more practical under conditions of produc- 
tion. This general thesis was concurred in and amplified by Messrs. Kunzman, 
Dubray, and Rackett. 

The topic of desirable subjects for meetings was then opened for general 
discussion by the Chairman. The following items were recommended : 

A description of the advances in and current status of television, as it is 
being developed by the local Don Lee station. 

A historical review of the development of the motion picture art at a meet- 
ing to be held at the Motion Picture Exhibit in the Los Angeles Museum, 
where Mr. E. Theisen, a member of the Society, is Honorary Curator. 

A demonstration and explanation of the projection test reel developed by 
the Projection Practice Committee. 


The developments in 16-mm. sound-on-film projection equipment. 
Technicolor's three-color process. 

The speaker of the evening, Mr. William Hartman of the Carl Zeiss organiza- 
tion, was then introduced by the Chairman. Mr. Hartman responded with an 
excellent dissertation on the planetarium now being constructed in Los Angeles 
on Hollywood Mountain, accompanying his descriptions with appropriate lantern 
slides. Mr. Hartman's talk was followed by an open forum of questions and dis- 
cussion which covered many phases of the optics and mechanics of the planetarium 
instrument, as well as some of the general facts of astronomy. 

The meeting was adjourned by the Chairman at 10:15 P.M. 


The first monthly meeting of the Section was held at the studio of RCA Photo- 
phone, Inc., New York, N. Y., on December 13th. First on the program was the 
presentation of a short motion picture, followed by a talk and demonstration by 
F. C. Barton, of the RCA Victor Company, on "High-Fidelity Lateral-Cut 
Disk Records." After the discussion of the presentation, another short subject 
was shown, which was then followed by an open forum discussion on the subject 
"Should Studio Recording Equipment Compensate for Theater Reproducing 


At a meeting held on December 14th at the studios of Burton Holmes Lectures, 
Inc., Chicago, 111., demonstrations and talks on the subject of laboratory practice 
were presented, as follows : 

"Airplane Racks," by T. L. Gibson, J. E. Brulatour, Inc. 

"Chemical Fades," by R. Tavenier, Mutual Film Laboratory. 

"Developer Exhaustion Tests," by H. Anders, Jam Handy Pictures Service, 

"Bloop Punch," by V. M. Bowers, Action Film Co. 

"Tight Winders," by R. F. Mitchell, Bell & Howell Co. 

"Moviola Sound Editing," by W. and H. Lignell, Burton Holmes Films, Inc. 

"Film Kinks," by V. Blakeley, Chicago Film Laboratory. 

"Filing Film," by E. Cour, Jeencour Productions. 


At a meeting held on December 6th at the General Office of the Society, the 
final form of the Revised Standards Booklet was decided upon, and final drawings 
are being made for publication in the JOURNAL in the near future. The Committee 
is also studying the practicability of establishing new dimensional standards for 
reel hubs. The Standard S.M.P.E. film perforation was adopted by the British 
Kinematograph Society, subject to acceptance as a Deutsche Industrie Normen. 
Proposals have also been submitted by the B. K. S. concerning a universal film 
core for all 35-mm. raw stock, which is now being considered by the Committee. 


TT rHAT big picture today does not in- 
clude backgrounds that call for com- 
posite photography? The answer is obvious. 
. . . The really vital point is: What medium 
to use in photographing these important 
backgrounds? . . . Eastman has answered that 
question. Eastman Background Negative, 
with its remarkably fine grain, its surprising 
speed, and its excellent processing charac- 
teristics, completely solves the film problem of 
the composite shot. Eastman Kodak Company, 
Rochester, N. Y. (J. E. Brulatour, Inc., Dis- 
tributors, New York, Chicago, Hollywood.) 

Background Negative 




Volume XXII FEBRUARY, 1934 Number 2 



Further Investigation of Ground Noise in Photographic Sound 
Records. . . .O. SANDVIK, V. C. HALL, AND W. K. GRIMWOOD 83 

Sound Film Printing II J. CRABTREE 98 

Recent Improvements in the Bell & Howell Fully Automatic 
Printer. . . . A. S. HOWELL AND R. F. MITCHELL 115 

The Economics of Projector Lamps for Advertising Purposes . . 

E. W. BEGGS 127 

Sixteen-Mm. Sound-on-Film J. O. BAKER 139 

Color for Industrial and Business Films 

R. H. RAY AND H. W. CRESS 144 

Some Practical Applications of Acoustics in Theaters 


Society Announcements 153 





Board of Editors 

J. I. CRABTREE, Chairman 


Subscription to non-members, $8.00 per annum; to members, $5.00 per annum, 
included in their annual membership dues; single copies, $1.00. A discount 
on subscriptions or single copies of 15 per cent is allowed to accredited agencies. 
Order from the Society of Motion Picture Engineers, Inc., 20th and Northampton 
Sts., Easton, Pa., or 33 W. 42nd St., New York, N. Y. 

Published monthly at Easton, Pa., by the Society of Motion Picture Engineers. 

Publication Office, 20th & Northampton Sts., Easton, Pa. 
General and Editorial Office, 33 West 42nd St., New York, N. Y. 
Entered as second class matter January 15, 1930, at the Post Office at Easton, 
Pa., under the Act of March 3, 1879. Copyrighted, 1934, by the Society of 
Motion Picture Engineers, Inc. 

Papers appearing in this Journal may be reprinted, abstracted, or abridged 
provided credit is given to the Journal of the Society of Motion Picture Engineers 
and to the author, or authors, of the papers in question. The Society is not re- 
sponsible for statements made by authors. 

Officers of the Society 

President: ALFRED N. GOLDSMITH, 444 Madison Ave., New York, N. 
Past President: JOHN I. CRABTREE, Kodak Park, Rochester, N. Y. 
Vice- President: WILLIAM C. KUNZMANN, Box 400, Cleveland, Ohio. 
Vice-President: OMER M. GLUNT, 463 West St., New York, N. Y. 
Secretary: JOHN H. KURLANDER, 2 Clearfield Ave., Bloomfield, N. J. 
Treasurer: TIMOTHY E. SHEA, 463 West St., New York, N. Y. 


EUGENE COUR, 1029 S. Wabash Ave., Chicago, 111. 
HERFORD T. COWLING, 7510 N. Ashland Ave., Chicago, 111. 
RALPH E. FARNHAM, Nela Park, Cleveland, Ohio. 
HERBERT GRIFFIN, 90 Gold St., New York, N. Y. 
EMERY HUSE, 6706 Santa Monica Blvd., Hollywood, Calif. 
WILBUR B. RAYTON, 635 St. Paul St., Rochester, N. Y. 
HOMER G. TASKER, 41-39 38th St., Long Island City, N. Y. 



Summary. This paper presents an investigation of the growth of the ground 
noise during the successive stages of the processes involved, from the time of manu- 
facturing the base support until the sound is reproduced in the theater, 

A study is made of the relation between emulsion grain size and the level of ground 
noise. When the processing and handling were done with the utmost care, the ground 
noise was found to decrease continuously as the emulsion grain size decreased. 

Finally, a study was made of the character of the noise from the standpoint of the 
relation between its r.m.s. value and its average and peak values, as well as a de- 
termination of the noise as a function of frequency. 

In two earlier papers 1 on this subject, the origin of ground noise was 
investigated, and attempts were made to reduce it by certain types of 
surface treatments. The aim of the present paper is to determine the 
growth of the noise level during the successive stages of the processes 
involved from the time of manufacturing the base support until the 
sound is reproduced in the theater, and to study its character in terms 
of its r.m.s. value, its peak values, and the distribution of its energy 
as a function of the frequency. 

In the first paper on the subject, data of a rather qualitative nature 
were published, showing the growth of the noise during the various 
manufacturing and processing stages into the finished sound and 
picture. It was concluded at that time that the level of what might 
be called the inherent ground noise in the film was of a lower order 
than the level of the ground noise which was encountered in practice, 
that is, in the theater. The origin or cause of this high noise level 
was assigned to various kinds of surface damages, such as scratches, 
dirt, dust, finger-prints, and oil spots. It was also found that the 
noise continued to grow in a uniform and continuous manner with 
the number of times that the film was run through a projector. 

* Presented at the Fall, 1933, Meeting at Chicago, 111. Communication 
No. 525 from the Kodak Research Laboratories. 
** Eastman Kodak Co., Rochester, N. Y. 




These measurements have now been made with greater care and 
better facilities, and the results check our earlier conclusions in that 
the level of the noise encountered in practice, in general, is higher than 
the inherent or irreducible level of film noise present upon careful 
processing, and that it grows very rapidly as the film is run repeatedly 
through a projector. 

In view of these facts, it appears that the ground noise as en- 
countered in motion picture practice can not be lowered materially by 
decreasing what has been referred to as the inherent film noise. 


Growth of Ground Noise with Successive Stages of Manufacturing and 

Noise Level 

in Db. 


Film Base 
Film Base 

Positive Film 
Positive Film 

Positive Film 

Positive Film 

Positive Film 





Noise Level 
in Db. 




Fixed, washed, and dried 

Developed, fixed, washed, and 

Run through printer once, 
developed, fixed, washed, 
and dried 

Run through printer once 
with sound negative, de- 
veloped, fixed, washed, and 

Run through sound recorder 
once, developed, fixed, 
washed, and dried 












However, because of the tremendously important bearing which this 
question has on any sound recording process, it appears worth while 
to inquire what determines its irreducible limit. 

Accordingly, measurements were made to determine the actual 
noise level in the film at each successive stage of manufacture and 
processing. The results of these measurements are given in Table I. 
Columns 1 and 2 give the material and its treatment, respectively, 
while columns 3 and 4 show the noise level of the several materials 
and treatments, with and without the audibility network, respec- 

Feb., 1934] 



Before discussing these data it might be well to describe briefly the 
method of measurement and the function and characteristics of the 
audibility network. The measurements were made by running the 
film through a slightly modified sound head of a motion picture pro- 
jector with a scanning beam whose dimensions were 0.0008 by 0.084 
inch on the film. The modifications of the sound head were for the 
purpose of insuring low microphonic and system noises. Their sum, 
with maximum illumination on the photoelectric cell and the operat- 
ing gain of the amplifier as described below, was 63 db. to 64 db. below 
the level of the standard frequency record used. The combined over- 





FIG. 1. Curve showing the sensitivity of the ear as a function 
of the frequency at a loudness level of 30 db. ; also the attenuation 
in db. of the audibility network. 

all response of the photoelectric cell, the amplifier, and the measuring 
device was uniform over the frequency range of 50 to 10,000 cycles 
per second, and measured exactly the r.m.s. values of the current. 
The operating gain of the amplifier in all cases was adjusted by the aid 
of a variable width 1000-cycle record of about 80 per cent modulation. 
The gain was adjusted so that the output level of this record was 
+ 6J4 db. and all subsequent data were obtained using this as the 
reference level. The frequency characteristics of the system were 
checked frequently by the aid of a multiple frequency record set 
aside for that purpose. 


The sensitivity of the ear is not constant over the frequency range 
indicated, but varies with the frequency in some particular manner, 
which is different for different loudness levels. Therefore, in order 
that the physical measurements should correspond approximately to 
audition tests, it is necessary to be able to change the frequency re- 
sponse of the measuring system to conform to that of the ear at the 
loudness level in question, which was chosen as 30 db. above the 
audibility threshold. The change in the sensitivity of the ear with 
frequency at that loudness level is shown in Fig. 1, where the ordinates 
under the curve represent the increase in intensity in db. required at 
different frequencies for equal sensations. 

In order to change the frequency characteristic of the amplifier to 
conform to that of the ear, it was therefore necessary to insert a net- 
work, called an audibility network, which attenuated the current at 
the different frequencies by an amount equal to the ordinates under 
the curve in Fig. 1. 

Returning now to Table I, it is found that plain film base has a 
noise level of -43 db. and -39 db. (below the output of the 1000- 
cycle standard record), with and without the audibility network, 
respectively. This is the lowest one in the entire series. Film base 
gelatin-coated is next, and although there is little choice between 
film base gelatin-coated motion picture positive film after the silver 
halide is fixed out and the remaining gelatin-coated base is washed 
and dried. However, when positive film was subjected to the action 
of D-16 developer for 6 minutes before fixing; the noise increased 

In order to find the effect of printing, a sample of positive film was 
run through a printer once, a second sample was run through the 
printer once in contact with the sound negative, and a third sample 
was run through once in contact with the sound negative and a second 
time in contact with the picture negative. Each of these samples was 
then processed, that is, developed for 6 minutes in D-16 and fixed, 
washed, and dried. None of these three samples was found to be 
noisier than a sample of positive film taken directly from the original 
wrapper and processed. It is interesting to note, however, that a 
sample of positive film run once through the recorder and then proc- 
essed was found to be materially noisier than the sample run through 
the printer. This can be accounted for only by the fact that in the 
former case the sample was exposed to unfiltered, unconditioned air, 
while in the latter case the operation was carried out in a printing 

Feb., 1934] 



room where the air was filtered and conditioned. This serves to 
emphasize the extreme care with which film should be handled. 

The most pronounced increase in the noise during any one of the 
above steps occurs when the film is developed. This might be due to 
the development of a certain number of silver halide grains, that is, 
fog grains, which modulate the photoelectric cell illumination and, 


Growth of Ground Noise with Processing Treatment of Gelatin- Coated 
Film Base 

Material Treatment 

Film Base 

Gelatin-Coated None 

Gelatin-Coated Washed and dried 

Gelatin-Coated Fixed, washed, and dried 

Gelatin-Coated Developed, fixed, washed, and 

Noise Level 

in Db. 




-39 l / 2 

-38 1 /* 

Noise Level 

in Db. 






therefore, give rise to noise. It might also be due to suspended 
particles in the developer which are deposited on the surfaces and 
within the emulsion layer. 

In an attempt to determine which of these two suppositions is the 
correct one, tests were made on gelatin-coated base by subjecting 
it to the successive stages of processing. The results of these tests 


Growth of Ground Noise with Time of Development of Gelatin- Coated 
Film Base 

Time of Development 

in Minutes 

in D-16 





Noise Level in Db. 

with Audibility 



Noise Level in Db. 

without Audibility 



(Table II) show some difference in the noise at the successive stages, 
particularly when developed. It is not sufficiently great, however, 
to account for the increase in noise found in Table I. Further 
measurements were made on gelatin-coated base subjected to the 
action of D-16 developer for different lengths of time, as shown in 
Table III. These results indicate a gradual growth in noise with de- 



[J. S. M. P. E 

velopment time. However, with the audibility network particularly, 
this growth is not well defined ; further data, which will be presented 
below, show that the growth is due to a change in the grain charac- 
teristic with development. 

If the structure of a uniformly flashed image were optically homo- 
geneous, as the density of the deposit increases, thereby decreasing 
the photo-cell illumination, the noise level should decrease correspond- 
ingly. If the latter is not the case, then the microscopic structure 

I.O I* ZO 

Dirrusc. DENSITY 

FIG. 2. The two upper curves show the 
relation between ground noise and density of 
positive film developed in D-16 and D-89 de- 
velopers, respectively. The lowest curve 
shows the same relation calculated on the 
basis that the photographic deposit is opti- 
cally homogeneous and the illumination on the 
film constant. 

of the image, that is, the size and distribution of grains and grain 
aggregates, must be such as to modulate the photo-cell illumination. 
In order to determine whether the noise of a series of uniformly 
flashed and developed sound track densities decreased in the ratios 
of the corresponding decrease in the photo-cell illumination, samples 
of positive film were exposed to different amounts of light and de- 
veloped for 3 l /4 minutes in D-89 developer, giving a gamma of about 
0.6. A similar series of samples was prepared and developed for 6 


minutes in D-16 developer, that is, to a gamma of about 2.0. These 
samples were measured, and the results are shown in Table IV and 
Fig. 2. The relation between ground noise and density is more 
readily seen by reference to the figure. It is seen that for both sets of 
samples the noise actually increased with the density up to a diffuse 
density of 0.17, whence it decreased as the density increased. It 
should also be noticed that the noise level is higher on the samples 
developed in D-16 developer than it is for the corresponding densities 
on samples developed in D-89. 

It is evident from these data that the structure of the silver deposit 
itself has a very important bearing on the question of surface noise. 
This raises several questions : First, do the two developers mentioned 


Ground Noise as a Function of Density of Positive Film Developed in 
D-16 and D-89 Developers; Illumination on Film Constant 

Noise Level in Db. Noise Level in Db. 

with Audibility Network without Audibility Network 

Developed Developed Developed Developed 

6 Min. 3 1 /* Min. 6 Min. &/t Min. 

Density in D-16 in D-89 in D-16 in D-89 

0.03 -39 -41 -33 -34 

0.07 -36 -38 1 /* -32 -34 

0.10 -35 . -37V* -30V2 -32 

0.18 -33 1 /* -37 -29V 2 -31 1 A 

0.25 -33Vs -37Vs -29Va -32'A 

0.45 -37 -41V* -33 -35V* 

0.68 -40 -46V-2 -36 -39 

0.92 -44 -53Va -40 -44 

1.14 -49 -56 -45 -47 

1.50 -55 .. -50 

above produce images of fundamentally different microscopic struc- 
tures, or does the difference lie only in the different degrees of develop- 
ment? Second, in the case of the variable density type of records 
particularly, what combination of negative and print densities and 
gammas gives the greatest ratio of signal to noise? Finally, would 
there be any advantage from the standpoint of noise in using an 
emulsion of materially different grain characteristics? 

To investigate the relationship between the ground noise and the 
degree of development, a series of flashed and developed samples was 
prepared on positive film. The series of flashed exposures was timed 
so as to produce equal densities for the different lengths of time of 
development. The results of the measurements of these samples are 



[J. S. M. P. E. 

shown in Table V and Fig. 3. It is seen that the growth of noise with 
time of development in this case is much more pronounced than it 
was in the case of a similar series of tests on gelatin-coated film base 
developed for various lengths of time, the results of which were shown 
in Table III. 





FIG. 3. Variation of ground noise with gamma 
of positive film developed in D-16 developer. 

This growth of the noise is due apparently to the granularity of the 
silver deposit. Also, the differences in the values in columns 2 and 3 
or 4 and 5 in this table are evidently due primarily, if not entirely, to 
the difference in the degree of development. Unfortunately, no data 
were obtained on the noise from flashed exposures developed to equal 


Ground Noise as a Function of Gamma of Positive Film Developed in D-16 

Noise Level 

Time of 
in D-16 (Min.) 








in Db. 







Noise Level 

in Db. 









gammas in the two developers. The extrapolated curve in Fig. 2, 
however, should give a rough approximation of the noise at a gamma 
of 0.6 in D-16. This procedure indicates a noise level of about 7 
db., as compared to the observed value of 6*/2 db., at the same 
density and gamma, for the sample developed in D-89. 

Feb., 1934] 



Now, the second question, namely, which combination of negative 
and print density and gamma results in the greatest ratio of signal to 
surface noise, is a question which can not be properly treated without 
due consideration of the percentage modulation and other factors 
which affect harmonic distortion. Therefore, the few measurements 
which have been made bearing on that phase of the problem will not 
be included in this paper. 

The last question of the relation of the emulsion grain size and 
ground noise has been investigated at some length, and the results 







%VZ.E. <R<MN ElMUU. 


0-% 1.0 tb Z.O Z& 


FIG. 4. Curves showing the relative noise levels 
of three classes of emulsions, at different densities. 

which were obtained are quite encouraging. It is, of course, evident 
that as the discrete silver particles, which make up the developed 
image, are made smaller and smaller, the structure of the image 
should become more and more nearly optically homogeneous; until, 
in the limit, when the state of division is atomic, the structure would 
be continuous to the scanning beam. The granularity or the size of 
the grain aggregates in general is less in the photographic image 
developed from a fine-grain emulsion than it is from a coarse-grain 
emulsion. Therefore, if the size of the silver halide grains could be 
sufficiently reduced, the developed silver deposit should approach 



optical homogeneity. Whether or not this atomic state of division 
can be attained, it lies quite beyond the realm of practicability, owing 
to the important consideration of emulsion sensitivity. However, 
in order to get some idea of the relation between noise and grain size, 
samples were prepared from film coated with three classes of emul- 
sions, one with coarse grains, a second with medium sized grains, 


Comparison of Ground Noise as a Function of Density, of Three Classes 

of Emulsions; Measurements Made with Constant Illumination 

on Films 

Fine-Grain Emulsion 

Medium Size-Grain 

Coarse-Grain Emulsion 























0.20 ' 


























Noise Level 
in Db. 




Noise Level 

in Db. 







-33V 2 

-33 l /2 














-32 l /2 









-47V 2 

and a third with very fine grains. A series of uniformly flashed 
exposures was made in each class and developed in D-16 developer 
to gammas of 1.5, 2.0, and 3.8, respectively. The reason for not 
developing all three to equal gammas was that the maximum gamma 
of the coarse-grain emulsion was not much above 1.5, while the time 
of development for a gamma of 1.5 on the fine-grain emulsions was 
so short that it would have been difficult to develop it uniformly. 


The r.m.s. value of the total noise of these samples was determined 
as before, and the results are shown in Table VI and Fig. 4. The 
noise levels in each case are again expressed in terms of a level of 
+6*/2 db. for the standard 1000-cycle record. These data can be 
more readily compared by referring to Fig. 4, which shows the relative 
noise level of the three classes of emulsions for a series of specularly 
measured densities. The three curves are extrapolated to converge 
at the point represented by zero density and a level of 38 l /2 db., 
which point represents the noise level for developed clear film. The 
data show that the noise level decreases very decidedly with the 
grain size, the difference between the levels of the coarse-grain and 
the fine-grain ones at a specular density of 0.6 being nearly 12 db. 
This is encouraging, since when and as the processing and handling 
conditions improve so as to warrant it, finer and finer grained sound 
recording emulsions can be employed. Inasmuch as a decrease in 
the grain size is accompanied by a corresponding sacrifice in the 
emulsion sensitivity, and since the adoption of sound recording emul- 
sions with materially finer grains than the present sound recording 
emulsions would probably also necessitate considerable change in the 
processing, it is the opinion of the authors that, so far as the motion 
picture theater field is concerned at present, such a change is not 
advisable; particularly since the size of the grain, and, therefore, 
the granularity of the photographic image, contributes only a small 
fraction of the noise levels which obtain in practice. 

Some additional data were obtained which deal more particularly 
with the character of the noise, from the standpoint of types and their 
origin. That part of the investigation consisted in determining the 
number of voltage peaks lying above successively increasing levels, 
and in determining the distribution of noise as a function of the 

The description of the apparatus used for these analyses will be 
omitted, except to state that the apparatus for determining the num- 
ber of peak voltages consisted of a number of recording peak reading 
voltmeters. The recording was accomplished by means of Central 
Scientific Co. impulse counters. These counters were calibrated so as 
to trip on voltage peaks corresponding to r.m.s. values of +24, +28, 
and +32 db., respectively. The peak voltmeters used were of the 
full- wave type and, therefore, operated the counter on voltage im- 
pulses caused by either a sudden increase or a sudden decrease in the 
opacity. By means of the volume indicator the gain was adjusted to 



[J. S. M. P. E. 

approximately equal levels for all densities. The volume indicator 
used reads very nearly average values. 

Its readings are given in the fourth column of Table VII, while the 
r.m.s. values are shown in the third column, and the number of voltage 
peaks, corresponding to r.m.s. values of +24, +28, and +32 db., 
are shown in columns 5, 6, and 7, respectively. Each counter was 
capable of recording peaks at the rate of 450 per minute. Thus, a 


Distribution of Noise in Terms of Number of Peak Voltages above Given 
R.M.S. Levels 

Noise Level in Db. 
without Audibility Network 
Number of Peaks per Min. 

above +28 above +32 

Sample Density 



above -f 

Plain Film Base 








Film Base 






















Medium Size- 





Grain Emulsion 






















































































value in excess of 440 indicates at least one voltage pulse of the 
indicated height every 1 /8 second. 

It should be remembered that in this case the gam of the amplifier 
was not adjusted by means of the standard record; that is, the levels 
in Table VII have no particular relation to the level of +6 x /2 db. for 
the standard record, although the relation could, of course, be ob- 
tained from the record kept of the gains which were used. The pur- 

Feb., 1934] 



pose was to get some idea of the distribution of the voltage peaks 
with respect to some average value which was chosen arbitrarily to 
read at approximately +18 db. 

Two facts are readily observed from this table. First, the number 
of peaks above a given value fall off as the density increases. Second, 
the number of peaks at the highest level, particularly, are relatively 
fewer in number as the average value or the r.m.s. value of the so- 
called inherent noise increases. These facts, although not con- 
clusive, indicate rather definitely that the noise which contributes the 
bulk of the energy, that is, determines the r.m.s. value, has a different 








FIG. 5. Ground noise as a function of frequency, 
for different samples of film. 

origin from that which gives rise to large peaks indicated by loud 
bangs and crackles in the loud speaker. These latter are very prob- 
ably due to the surface conditions, that is, to dirt, dust, scratches, 
etc. It should, of course, be borne in mind that the amplifier gain for 
a level of +18 db. on the fine-grain emulsion is much greater than the 
gain for the same level at the same density on the coarse-grain 

Fig. 5 shows the distribution of the noise as a function of the fre- 
quency from 100 to 10,000 cycles per second, for the three classes of 
emulsions at a specular density of 0.6 and gammas the same as shown 


[J. S. M. p. E. 

in Table VI, and also for plain film base and gelatin-coated film base. 
The relative levels of the last two samples have been adjusted to equal 
illumination on the photo-cell with respect to the others. The rela- 
tive levels among samples are, however, not particularly reliable 
nor of any particular importance. The distribution of noise in any 
one sample shows no striking departure from the distribution of any 
other sample. The most marked difference perhaps is the gradual 
divergence of the curves for the three emulsions with frequency. One 
interpretation of this, which may or may not be correct, is that the 
high-frequency noise, in excess of that due to the film base, in general, 
is due to the grain structure, that is, the granularity of the photo- 
graphic image. 


> 1.6 

5> 1.4 
u i.o 





<XZ 0.4 O.fc 06 10 t.Z 1.4 1.6 I.* 1.0 

FIG. 6. Relation between specular (pho- 
toelectric cell) density and diffuse density 
for the three classes of emulsions. 

Fig. 6 is included to give the relation of diffuse to specular densities 
as used in this paper. The term "specular density" has been used 
for convenience it is actually photoelectric cell density as deter- 
mined in the sound head. 


1 CRABTREE, J. I., SANDVIK, O., AND IVES, C. E.: "The Surface Treatment of 
Sound Film," /. Soc. Mot. Pict. Eng., XIV (March, 1930), No. 3, p. 275; Kino- 
technik XII (1930), p. 390, 420. 

2 SANDVIK, O. : "A Study of Ground Noise in the Reproduction of Sound by 
Photographic Methods," Trans. Soc. Mot. Pict. Eng., XII (1928), No. 35, p. 790. 


MR. CRABTREE: It is interesting that the noise level-gamma curve at a con- 
stant density corresponds very closely to the graininess gamma curve at a constant 
density . 


In connection with the relatively large increase of the noise level as develop- 
ment progresses in certain developers, especially those containing a high con- 
centration of sulfite, the sulfite dissolves some of the emulsion which, in turn, is 
reduced to silver which tends to deposit on the film surface. If film that has 
been developed in such a solution is wiped with cotton, the cotton will remove a 
certain amount of this colloidal silver and rapidly become dirty. Also, developers 
of this type deposit a sludge of silver in the bottom of the tank and, if the de- 
veloper is agitated, the sludge becomes suspended again and tends to deposit on the 
film. The matter of developer filtration is, therefore, of some importance in this 

Have measurements been made on the effect of pure oil on the film? Some 
films are badly spattered with oil, and I should like to know whether that oil has 
any effect on the ground noise. I don't mean oil mixed with dirt; I mean pure 
clean oil on the film. 

MR. SANDVIK: If the oil is uniformly distributed over the surface of the film, 
it usually decreases the ground noise, particularly after the film has been run 
through the projector a number of times. If it is patchy, then it causes a certain 
amount of low-frequency noise, due to the large patches, but not very much. 
It amounts to only about 2 db. 

MR. CARVER: Has the noise that is measured any relation at all to the noise 
we actually hear? What I hear in a film is a scratching that can't possibly have 
anything to do with a very fine deposit of colloidal silver. Occasionally there will 
be a little clicking noise. This is what I believe every one hears when ordinary 
film is projected too loud. It seems to me that the colloidal silver would produce 
a uniform hissing noise which I probably couldn't hear at all. Is that possible? 

MR. SANDVIK: If you listen to a film with a uniform density, which has been 
processed and handled very carefully so as to have no large scratches or large dust 
particles on it, I think that you will find that the noise is more a hissing noise than 
anything else. But, of course, after the film has been run through the projector 
once or a dozen times, it suffers considerable surface damage, which causes what 
you probably describe as scratches or bangs and booms. Those constitute an 
annoyance factor to the ear which can not be evaluated by physical measure- 
ments alone. 

MR. CARVER : Those are really the noises that matter, aren't they? 

MR. SANDVIK: The other, that is, the continuous hiss, is important too; al- 
though at the present time it is probably not of great importance from the stand- 
point of the theater because the surface damage raises the noise level to an extent 
such that what I call the inherent ground noise in a film is inherently small. 

There is another thing that has to be taken into account, and that is that even 
though the distribution of the noise is fairly uniform and slopes up somewhat at 
the low frequencies, the sensitivity of the ear decreases so tremendously at the 
lower end that you aren't conscious of as great a disturbance at those frequencies. 

MR. CRABTREE : What is the magnitude of the ground noise of cellulose acetate 
records as compared with that of good film records? 

MR. WILLIS: Possibly 10 db. lower in the best cases. 


Summary. The production of sound-film prints from variable density negatives 
by the Model D Bell & Howell printer has been studied from the point of view of high- 
frequency response and uniformity of product. The account of this study, begun in 
Part I 1 , is continued here, with particular reference to the degree of influence of 
slippage on the high-frequency response, occasioned particularly by non-conformity 
of the perforation pitch of the negative and positive films. It is found that to improve 
printing conditions in practice, it is first necessary to achieve consistency in the pitch 
of the processed negative and positive materials and to make the pitch of the proc- 
essed negative 0.0004 inch less than that of the positive raw stock. 

In an earlier paper 1 the author referred to the small values of shrink- 
age that result from processing present-day motion picture film. In 
the sound picture laboratory of the Bell Telephone Laboratories, 
where processing machines representative of those used in commercial 
laboratories are employed, the average value of shrinkage resulting 
from processing is about 0.025 per cent. 

The Bell & Howell Model D continuous printer is widely used in 
sound film printing. Printing is effected in this device at the pe- 
riphery of a sprocket of such curvature as to accommodate, without 
slippage, a negative that is 0.368 per cent below standard pitch and 
positive raw stock that is 0.079 per cent below the standard pitch of 
0.187 inch. The following discussion is based upon studies made 
with that printer. 

Assuming that the negative and positive raw stocks are full pitch, 
the negative, if printed soon after processing, will be approximately 
0.3 per cent oversize for the sprocket, and slippage of that amount 
between negative and positive will occur at the printing aperture. If 
the pitch of the negative raw stock is above standard, as occasionally 
occurs, or, if the positive raw stock is undersize, which we have found 
to be usually the case, the amount of slippage is increased by the ex- 
tent of oversize or undersize, as the case may be. The effect of slip- 
page between negative and positive during the printed exposure is 

* Presented at the Fall, 1933, Meeting at Chicago, 111. 
** Bell Telephone Laboratories, New York, N. Y. 



to cause a blurring of the printing image, which results in a reduction 
of amplitude of the recorded signal to an extent depending upon the 
recorded frequency, being negligible at low frequencies but very appre- 
ciable at high frequencies. 

The manner and degree of this degradation of print definition de- 
pends upon the height of the printing aperture and to some extent on 
the sprocket tooth contour. 

Some idea of the degree of the influence of slippage on high-fre- 
quency response can be obtained theoretically by calculation if certain 
assumptions are made. Such a calculation has been made on the 


? 6 

9 a 

0.0006 0.0004 0.0002 

0.0004 0.0006 


FIG. 1. Calculated printing loss at 9000 cycles due to slippage be- 
tween negative and positive. 



FIG. 2. Influence of film pitch dimensions on manner of propulsion 
by sprocket. 


J. CRABTREE [j. s. M. P. E. 






FIG. 3(A). Effect of pitch measurement of negative on print uniformity; 9000 
cycles: printed on positive stock of pitch 0.1872 in. 

basis of two such assumptions: first, the use of an aperture height 
equal to twice the perforation pitch ( 3 /s inch) and, second, that slip- 
page, when it occurs, is practically instantaneous. Under such con- 
ditions the resulting print may be regarded as approximately equal to 
that of two sine waves of transmission which are out of phase by an 
amount equal to the degree of slippage as the load is transferred from 
tooth to tooth. Computation of the amplitude loss at a frequency 
of 9000 cycles resulting from different values of slippage produces the 
curve shown in Fig. 1. On the left-hand side of the peak, the nega- 
tive is oversized ; on the right-hand side, undersized. Conditions have 
been encountered in normal printing practice where slippage of 
0.0006 inch or more has occurred. The computed loss from such a 
condition is shown to be nearly 5 db. at 9000 cycles. It was thought 

Feb., 1934] 









FIG. 3(B). Same as Fig. 3(A): pitch of positive stock 0.1870 in. 

important to determine experimentally the magnitude of the losses 
resulting from mismatching the films and sprocket since, it should be 
noted, slippage is not the only undersirable factor resulting from pitch 

It is well known that when a film is oversize for the sprocket, it 
is driven by the pair of entering teeth, as shown in Fig. 2(C). In the 
case of the printer under discussion, the entering teeth precede the 
printing aperture, so that the film is no longer under tension at the 
printer gate but is being pushed downward between the aperture plate 
and the shoe. The tendency of a film driven by the entering teeth 

[J, S. M. P. E. 




FIG. 3(C). 
FIG. 3(C). Same as Fig. 3(A): pitch of positive stock 0.1868 in. 

is to ride up on the face of the tooth away from the shoulder of the 
sprocket. This tendency will probably be increased by friction in the 
gate, and opportunity for a loss of contact between the negative and 
the positive results. Examination of microdensitometric traces of 
high-frequency envelopes printed in this manner indicates that such 
conditions occur. 

In Figs. 3 (A), 3 (B), 3(C), and 3 (D) are shown traces of sections, three 
frames in length, of 9000-cycle records printed from negatives having 
different pitches on positive raw stocks of different pitches. Constric- 
tions in the wave envelopes indicate losses of modulation due to the 
loss of contact between the negative and the positive during printing. 
It will be seen that such losses are greatest when the negative approxi- 
mates full pitch, but that the losses decrease and finally disappear as 
the negative shrinkage attains a certain value. It shou!4 also be 

Feb., 1934] 








FIG. 3(D). Same as Fig. 3(A) : pitch of positive stock 0.1866 in. 

noted where the negative is oversize, that the uneveness of the en- 
velope is less when printed on full-pitch or oversize positive than 
when the print is made on undersize positive. That is, the worst 
condition for contact losses occurs when the negative is oversize and 
the positive undersize, a condition common in practice. 

When both the negative and the positive are oversize, and, there- 
fore, not under tension at the aperture, they ride in and out on the 
teeth more or less in contact; whereas, if the positive is shorter, or 
undersize, and, therefore, under tension, the underlying oversize nega- 
tive can not ride out on the teeth: it therefore buckles, causing a 
loss of contact and the consequent loss of amplitude, as shown. 

The study of a large number of microdensitometric traces of prints 
made under different conditions suggested that in the case of printing 
with oversize raw stock the irregularities tend to be less when the 
perforations are similar in both the negative and the positive than 




[J. S M. P. E. 






FIG. 4. Effect of perforation shape on print uniformity; 9000 cycles. 

when they are dissimilar, as is the case in commercial practice 
(Fig. 4). This appears to be reasonable, since the dimensions of the 
perforation probably determine the behavior of the oversize film in 
relation to the sprocket teeth. The difference in size and shape of 
the two types of perforations now in use (Fig. 5) is apparently suf- 
ficient to cause a difference in the behavior of the films as they pass 
through the printer. 

It was further observed, on comparing the prints made by a printer 
using the original printing sprocket with those made by a printer 
using the more recent sprocket which is made to closer manufactur- 
ing tolerances, that the matching of the pitch of the negative to that of 
the sprocket is of greater importance than is the presence of slight er- 
rors in the sprocket (Fig. 6), while still recognizing the fact that 







FIG. 4. (Cont.) 

sprocket errors are probably the cause of the irregular separation of 
the films. This latter fact is evident from Fig. 7, which shows the 
traces of a 9000-cycle print representing the same twenty teeth of a 
sprocket for three successive revolutions. The periodicity of the 
constrictions of the envelope with rotation of the sprocket shows that 
the print defects result from sprocket error. 

We have, then, the following conditions affecting printing loss and 
uniformity : 

(1) Slippage between films, resulting from mismatching of negative and posi- 
tive pitches, and causing loss of modulation by phase displacement of the over- 
lapping printed images. 

(2) Loss of contact between films due to buckling consequent on mismatching 
of negative or positive, or both, with the printing sprocket. This causes a loss of 
definition of the printed images at the points of buckle, resulting, in the print, in 
an irregular amplitude loss and non-uniformity of wave envelope. 



[J. S. M. P. E. 

To obtain a measure of the sum of these losses, negatives were re- 
corded on films covering a relatively wide range of pitch dimensions 
and having both types of perforation, and were printed on positive 
film, also covering a relatively wide range of pitch dimensions and 
of both types of perforation. The reason for examining the various 
combinations of the two types of perforation was, of course, to estab- 
lish a preference for a given combination if a physical basis for such 
preference should be found to exist. 

The negatives used were made in two ways : 

(1) Constant frequency records were made on a sufficient length of full- 
pitch material. After processing, a section of the negative was removed and 


FIG. 5. Showing differences in dimensions of negative and positive 

kept in a small sealed container while the remainder was left in the drying cabi- 
nets of the processing machines, where it was subjected to a current of air at a 
temperature of about 100F. Pitch measurements were made each day, and as 
the desired degrees of shrinkage were attained, sections were removed and stored. 
When the extreme degree of shrinkage desired had been attained, the various 
sections were assembled, and after storing the assembly for a few days to permit 
the whole to assume a uniform moisture content, the assembled negative was used 
for printing. 

(2) From a series of emulsions of raw film stock purchased over a period of 
two years it was possible to obtain a range of pitch values from 0.18715 to 0.1862 


inch in the same type of perforation. Sections of the various stocks were spliced 
together, and 9000 cycles recorded and the film processed as one unit. The fre- 
quency response of the various sections of the negative was measured to insure that 
no differences in recorded amplitude existed; any sections showing such differ- 
ences were excluded from the experiment. 

Of the stocks used for making the prints, some were selected from 
new consignments of film as received, six items were specially ordered 
for specific values of pitch, and the lowest pitch values were obtained 
from our older emulsions referred to above. Measurements of re- 
solving power of each emulsion were made by static printing in a 
printing frame from a closely spaced line grating test object and mea- 
suring the amplitude of the resulting wave envelope to insure uni- 
formity in this respect. 

The assembled negatives were then printed in the usual manner 
in the printer on each raw stock in turn. The pitch of each section 
of negative was measured before and after each printing, while the 
raw stock on which the print was made was measured at the begin- 
ning and end of the print. A length of about one hundred feet was 
discarded from the outside of each roll to avoid any possible error from 
drying out of the outer turns. 

Pitch measurements were made by laying the film on a flat surface 
and measuring the length of one hundred perforations with a high- 
quality steel rule graduated in hundredths of an inch. 

The prints were processed in a continuous machine to an over-all 
sensitometric gamma of 1.0. The response of the prints was measured 
on a Western Electric re-recording machine, using equalization. The 
frequencies were 9000 cycles and 2000 cycles, recorded on opposite 
sides of the negative. The loss of 2000-cycle output due to slippage 
was assumed to be negligible, the difference between the 9000-cycle 
and 2000-cycle levels being taken as the 9000-cycle loss. The volume 
indicator was a vacuum tube voltmeter with a damped meter circuit 
measuring the r. m. s. output. Its readings checked those made with 
a thermocouple. 

The 9000-cycle loss was plotted against the pitch of the negative 
for each positive that was printed. The curve so obtained gave the 
9000-cycle loss as a function of the negative pitch for a particular 
positive pitch and type of perforation. Several prints were made on 
different occasions and the results averaged, so that each curve is 
derived from a considerable number of measurements. All curves 
in a given combination of perforation types are assembled on one 

108 J. CRABTREE [j. S. M. P. E. 






....- - - 




FIG. 6. Effect of film pitch dimensions and sprocket accuracy on print 
uniformity; 9000 cycles. 







n i o.^^ 

FIG. 6. (Cont.) 



[J. S. M. P. E. 

diagram in order to show the ef- 
fect of the positive pitch as well 
as the negative for that combina- 
tion. The combinations are: 

Fig. 8: negative perforations printed 

on negative perforations. 
Fig. 9: negative perforations printed 

on positive perforations. 
Fig. 10: positive perforations printed 

on negative perforations. 
Fig. 11: positive perforations printed 

on positive perforations. 

It is obvious from the figures 
that for any positive material 
there is an optimal value of pitch 
for the negative for which the 
9000-cycle loss is least, and that it 
is less than that of the positive by 
an amount that is fairly constant. 
In Figs. 12 to 15 the curves are 
replotted, showing the 9000-cycle 
loss as a function of the difference 
between the negative and the posi- 
tive pitch. These curves show that 
within the error of experiment the 
optimal difference between the 
negative and the positive is the 
same, at least from the range of 
positive pitch likely to be en- 
countered in practice. 

In Fig. 16, each group is 
merged into one curve, enabling 
a comparison of the four group 
averages to be made. These, in 
turn, are reduced to an average 
curve in Fig. 17. The latter curve 
shows that the averaged optimal 
difference of pitch between the 
negative and the positive as de- 
termined from these experiments is 0.000425, or 0.23 per cent of 
the standard pitch value. The theoretical value derived from the 

FIG. 7. Envelope of 9000-cycle 
print, showing periodicity of varia^ 
tions with sprocket rotation. 

Feb., 1934] 



01872 01870 O 1868 O 1866 01864 01862 


FIG. 8. Printing loss (at 9000 cycles) as a function of negative 
pitch, for positives of pitch dimensions as shown: negative with 
negative perforations; positive with negative perforations. 

FIG. 9. Same as Fig. 8: negative with negative perforations; 
positive with positive perforations. 

FIG. 10. Same as Fig. 8: negative with positive perforations; 
positive with negative perforations. 

FIG. 11. Same as Fig. 8: negative with positive perforations; 
positive with positive perforations. 



[J. S. M. P. E. 

0.0002 0.0004 00006 00008 00010 


FIG. 12. Printing loss (at 9000 cycles) as a function of difference 
in pitch between negative and positive, for positives of pitch dimen- 
sions as shown : negative with negative perforations ; positive with 
negative perforations. 

FIG. 13. Same as Fig. 12: negative with negative perforations; 
positive w'th positive perforations. 

FIG. 14. Same as Fig. 12: negative with positive perforations; 
positive with negative perforations. 

measurement of sprocket diameter and the film thickness is 0.00054 
or 0.29 per cent of the standard pitch value. No explanation is 
advanced at this time for the discrepancy, which is small compared 
with the pitch variations occurring in film. 

It will be seen from the final average curve that the slopes on either 
side of the peak are approximately the same; that is, the loss for a 

Feb., 1934] 







00004 00002 0.0002 00004 0.0006 00008 00010 


FIG. 15. Same as Fig. 12: negative with positive perforations; 
positive with positive perforations. 

FIG. 16. Printing loss (at 9000 cycles) as a function of difference 

in pitch between negative and positive ; averages from Figs. 12 to 
15, inclusive. 

Negative Positive 

(A) neg. neg. 

(B) neg. pos. 

(C) pos. neg. 

(D) pos pos. 

FIG. 17. Printing loss (at 9000 cycles) as a function of differ- 
ence in pitch between negative and positive ; average over-all curve. 

given slippage between the negative and the positive is the same 
whether the slippage is plus or minus, although microdensitometric 
records show that on the oversize side of the optimum, the print 
envelopes are not uniform, whereas on the undersize side, the en- 
velopes are smooth. 


The group averages as in Fig. 16 show an apparent but slight pref- 
erence for negatives with Bell & Howell perforations. That is, how- 
ever, not entirely in accord with the observations on envelope uni- 
formity on page 103. 

The connection between these results and commercial practice 
is of considerable significance. The results show that under the con- 
ditions of processing employed in this laboratory, which are regarded 
as representative of commercial practice, the use of negative and 
positive raw stocks of standard pitch will result in a printing loss of 
at least 3 db. at 9000 cycles when printing is performed soon after 
the negative is developed. With positive stocks below the standard 
pitch, the loss mentioned above is increased by amounts that can be 
deduced from the curve in Fig. 16, and the non-uniformity of the wave 
envelope is aggravated. 

To improve printing conditions in practice, it is first necessary to 
achieve consistency in the pitch of the positive and negative materials. 
Next, the pitch of the processed negative should be approximately 
0.0004 inch less than that of the positive raw stock. If a survey of 
commercial processing conditions shows that the average negative 
development does not reduce the negative pitch to that value, sub- 
standard perforation of the negative raw stock should be resorted 
to. Similarity in the types of perforation of the negative and the 
positive may insure some improvement in print uniformity where 
oversized film is printed, and the pitch values of the negative and 
positive must match that of the sprocket. 


1 CRABTREE, J.: "Sound Film PrintingI" J. Soc. Mot. Pict. Eng., XXI 
(Oct., 1933), No. 4, p. 294. 



Summary. The results of a year of experience with the engineering model of 
this printer in actual production printing are outlined. The actual constructional 
changes that have been made in the printer are described and illustrated, and new 
developments that enable prompt and accurate sensitometric control to be maintained 
under production conditions are outlined briefly. 

The engineering model of the fully automatic sound and picture 
printer described previously 1 has been in continuous use at the 
M-G-M laboratory at Culver City, Calif., for more than a year. 
During that time complete and exhaustive tests were made that 
amply demonstrated the success of the design. It was operated on 
an average of twenty or more hours a day so successfully that a 
battery of printers was ordered. This battery, just completed, is 
shown in Fig. 1. Furthermore, experience with the new printer 
has enabled the M-G-M laboratory to utilize many refinements of 
sensitometric control of processing developed by the Bell & Howell Co. 


Some idea of the performance of the printer may be gathered from 
the experience of the M-G-M laboratory. One of the early tests 
involved printing a six- or seven-reel picture. One reel was printed 
on the new printer while the remaining reels were printed in the usual 
way. After quite a few prints had been made it was found that the 
new printer afforded such superior "snap" to the picture and quality 
to the sound that the one reel contrasted boldly in comparison with 
the rest. As a result, the remainder of the picture was printed on the 
new printer. The rush of work that followed showed what the 
machine could do in an emergency. 

Recently, an NRA subject had to be printed in a hurry. One 
thousand prints were made without removing the negative for clean- 
ing; at the end of the run the negative and matte films were in 

* Presented at the Fall, 1933, Meeting at Chicago, 111. 
** Bell & Howell Co., Chicago, 111. 




perfect condition and fit to make another run. The outstanding 
nature of this performance is obvious when it is realized that in 
ordinary practice the negatives must be removed after every few 
prints in order to be cleaned. Not only did the machine avoid the 
delays of cleaning, but the resultant wear and tear on the negative, 
as well. Furthermore, there was no waste due to breakage, mislights, 
asynchronism, etc. The machine was operated 24 hours a day until 
the complete release was finished. All the operator had to do was 
to feed positive stock into it. 

FIG. 1. Battery of new fully automatic printers ready for operation. 
(Note rheostat unit just below voltmeter.) 

Such examples are typical, and indicate the accomplishments 
possible with the new instrument. It is obvious that a year's test, 
involving every contingency and difficulty that could be imagined, 
has indicated a number of valuable refinements in the original design. 
Following is a brief outline of the various changes effected in the 
final model. 


The new machine is noticeably thinner and appears more compact 
because of its smoother outline. Refinements in gear cutting have 
made its operation even quieter than before. 


Main Sprocket. A change of major importance involves the shape 
of the sprocket teeth and the angle of the arc of contact of the films 
at the aperture. The shape of the tooth departs slightly more than 
that of the model D tooth from the true involute curve, and is de- 
signed to provide a straight-line wedge action so that the film, within 
the range of shrinkages now encountered, is transferred smoothly 
from one tooth to the next without jerking. 

It was found necessary to go to extremes in making the main 
printing sprockets of the required accuracy. A special hobbing 
machine, fitted with specially made hobs, is used exclusively for 
generating the main printing control sprockets. Each tooth is 
individually finished, and then the tooth contour is checked in a 
contour projector and the pitch is checked on an optical dividing 
head. The sprockets are now made from stainless steel, thus assuring 
a permanently clean, smooth tooth surface. As a result, the con- 
trolling dimensions of tooth radius, pitch, and other critical factors 
are held to a tolerance of 0.0002 inch. 

Because of the smoother operation of the machine and of using a 
printing aperture considerably smaller than that of the model D 
printer, the angle of contact of the films at the aperture has been 
reduced from approximately 41 to 24 degrees for the negative, and 
from approximately 32 to 19 degrees for the positive. This still 
brings the positive and negative films to perfect coincidence just 
before they reach the aperture, and allows the films to be stripped 
off the sprocket immediately after the films pass the aperture. This, 
in conjunction with the air pressure on both sides of the film at the 
aperture, reduces wear and abrasion of the films, an important factor 
in achieving a long life of the negative. 

All the factors which were investigated in this connection will 
not be discussed; suffice it to say that the combination of the new 
tooth shape, precision of construction, and the change of the arc 
of contact of the films at the aperture combine to produce the kind 
of work mentioned above. Incidentally, due to the ability of the 
machine to operate in both directions, the wearing of the sprocket 
teeth is equalized, and twice the ordinary effective life is attained as 
compared with the usual case where the machines operate only in 
one direction. 

Changing Parts. Another important improvement is the arrange- 
ment that allows important parts to be changed quickly. 

(a) The aperture plate. This can be removed in a few seconds 



for cleaning, polishing, or replacement by another type (for example, 
for 17V2-mm. sound negative). It is necessary only to loosen one 
locking screw. 

(b) The main sprocket. The complete flywheel unit is remov- 
able by unscrewing the six bolts that hold it to the main frame. This 
exposes the outer side of the sprocket, as in Fig. 2. The aperture plate 

FIG. 2. Close-up of right-hand head, with mechanical filter removed 
to show lower part of interchangeable aperture plate and ease of removing 
sprocket ; also the method of adjusting the tension of the tension rollers by 
moving along the slide the set screw holding the roller. The white pointers 
are set between the two index marks when threading to place the correct 
tension on the films. 

is removed, and then the sprocket and the light trap ring can be re- 
moved. This construction permits the main sprocket to be replaced 
at any time by the laboratory mechanic, with the assurance of re- 
taining perfect alignment. 

(c) Roller assemblies. The rollers are made with self-contained 
ball bearings permanently retained in position by an expansion of the 
ends of the inner bearing tube. This allows rollers to be changed 

Feb., 1934] 



instantly without danger of losing the balls and bearings, and yet 
with assurance of maintaining perfect concentricity. 

Optics. The optical system has been improved in several respects. 
First, the reflector was eliminated, as it was found to introduce com- 
plications: if the lamp were moved for any reason, the reflector had 
to be reset, with the possibility of affecting the exposure value of the 
light. By re-designing the condensers and other optical parts the 
effective even light of the former arrangement was more than 
doubled, despite the removal of the reflector. Furthermore, the 
exposure is comparatively unaffected by slight changes in the posi- 

FIG. 3. 

Diagram of the improved optical system giving greatly in- 
creased but very even illumination. 

tion of the lamp due to imperfect seating in the holder, etc., yet the 
illumination over the entire printing aperture is exceptionally uni- 
form. For instance, a fog test made on the machine can be drawn 
past the eye without showing evidence of unevenness. Such a test 
is decidedly more sensitive than any measurement of the variation 
of density on any densitometer in regular use. 

The optical system has been arranged to provide a moderate degree 
of diffusion of the light. Extensive tests showed conclusive evidence 
that purely specular illumination introduced too many complications, 
reproducing minute scratches on the celluloid side of the picture and 

120 A. S. HOWELL AND R. F. MITCHELL [J. S. M. P. E. 

sound negatives, whereas a moderate degree of diffusion eliminated 
them for all practical purposes. The evenness of the illumination, 
and the narrow printing aperture and perfect smoothness of film 
travel past the aperture, were the factors of principal importance 

Evenness of Operation. It is evident that the fog test is also quite 
critical with respect to indicating slight variations of the speed of the 
printer. This is especially so considering that the picture printing 
aperture has been reduced from 6 /ie to 3 /ie inch, and the sound print- 

FIG. 4. Close-up of rear of 
rheostat unit, showing rheostat 
handle interlocked with the water 
supply, and motor reversing switch 
on the starting handle. 

ing aperture from 5 /i 6 to 3 /32 inch. This reduction has been made 
possible by the improvements of the optical system and by the 
accurate speed regulation achieved by the combination of synchronous 
motor, precision gears, flywheel, and mechanical filter for operating 
the printer sprocket. The smaller the printing aperture the higher 
the sound frequency that can be printed, the "crisper" the print, and 
the more difficult it is to make a printer that will function perfectly. 
Rheostat Control. In place of the master control matte originally 

Feb., 1934] 



fitted, the printer is now furnished with an alternative rheostat con- 
trol as illustrated in Figs. 4 and 5. The rheostat is divided into three 

The main rheostat is operated from the rear (Fig. 4). Its purpose 
is to reduce the initial shock of switching 1000 watts directly on the 
same line that feeds the other printer lamps, and to avoid surges 
that might affect the other printers on the line. Separate combina- 
tion rheostats in front are employed to adjust the voltages of the two 
printer lamps very precisely. Each of the two setting dials is com- 
prised of an inner and outer unit, each of which is calibrated in twenty 

FIG. 5 (a). The front of the rheostat with cover open, showing double 
dials and trip plunger. The outer part of each dial has voltage steps of 2 
volts, the inner of 0.2 volt. 

steps, the outer dials in steps of 2 volts, and the inner dials in steps 
of Vio volt (Fig. 5(a)). 

This arrangement allows minute adjustments of the printing lamp 
voltages to be made with precision, to follow the variations of the 
speeds of different emulsions. When the control rheostats are 
adjusted for a given emulsion and given developing conditions, the 
voltages are marked on the cards on the rheostat cover (Fig. 5(6)). 
The set-up man then locks the cover and, therefore, has sole responsi- 
bility for the printer light setting. The arrangement allows the 

122 A. S. HOWELL AND R. F. MITCHELL [j. s. M. P. E. 

operator to start the machine at any time by turning on the main 
rheostat at the rear. 

The main rheostat lever is interlocked with the water supply so 
that the water is automatically turned on when the lamps are turned 
on. The rheostat coils are immersed in transformer oil; the entire 
unit has a water jacket for cooling. 

Printing Lamps. Improvement has been effected in the printing 
illumination by utilizing special 115-volt, 500- watt lamps of the 
on-course Beacon type, of the T-20 bulb size with standard mogul 
prefocused bases. Normally the lamps are operated at about 85 to 

FIG. 5 (&). The front of the rheostat with cover closed and locked. The 
cards are for lamp voltage records. Between the cards is the emergency 
trip (electrical) . Above is the voltmeter switch for reading the voltage on the 
sound head, line, and picture head, respectively. 

95 volts, so that they will last a year or two. The main advantage 
of operating at such a low level is that the danger of destroying the 
filament is practically eliminated and the illumination is maintained 
practically constant. 

The decrease of illumination due to bulb blackening is so slight 
that it is not noticeable in operation ; the rheostat settings are changed 
often enough to take care of other variations, so that the slight 
decrease due to that cause is compensated for automatically. 

Mattes. The control of the printer depends to a large extent on 
the control in making the mattes. It was found that unless the 
density of the opaque parts of the matte were kept above a certain 


value, enough light was passed to affect the efficiency of the narrower 
mattes. Increasing the exposure and development time to achieve 
the requisite density introduced the complications of the Eberhard 
effect: there was enough "burn over" into the clear portion of the 
matte to reduce its effective transmission value, apart from the fact 
that it was difficult to keep the base fog down to a minimum. 

The problem was solved by using a special experimental high- 
contrast emulsion, and maintaining the density of the dark part of 
the matte negative at a value of 1.65-1.90 and of the positive matte 
2.0-2.2. No determination of gamma is made, as the development 


FIG. 6. New matte film matching unit installed on regular B&H 
splicing machine. 

occurs in the usual positive developer under conditions that afford 
a normal gamma of 2.0 with regular positive stock. 

The widths of the various matte openings were revised to provide 
an even geometric progression of exposure at the aperture, using 
mattes of the above-mentioned transmission characteristics in con- 
junction with the improved optical system. The changes in the 
mattes required only that a new set of matte slides be made. 1 How- 
ever, a more convenient method of matching the mattes to the nega- 
tives has been devised. The original method involved running the 
negative on the splicer at the same time as the mattes were being 
spliced. In the present method, a careful record of the scene lengths 
of the negative, correct to the frame, is made. 



An attachment that can be fitted in a few minutes to any splicer is 
used to measure the matte film to the corresponding lengths (Fig. 6) . 
Inasmuch as the matte length is one quarter that of the negative, the 
splicing attachment is geared accordingly, but the footage dial and 
frame counter register as feet and frames of negative. An integral 
punch permits locating the exact perforation at which the mattes 
are to be spliced, so that the complete matte can be made up without 
touching the negative. 

FIG. 7. Switchboard unit with cover removed. Note valves on air 
inlets operated by solenoid at bottom center. The three tumbler switches 
immediately above operate the three edge-printing lamps. 


Air Valve. An air valve has been arranged to interlock with the 
regular tripping mechanism so that the machine will stop if the air 
supply fails. Valves have also been located at the main air inlet, 
which open when the machine is started and close when it is stopped. 
These can be seen clearly in Fig. 7. 

Electrical. The machine is operated by a 220-volt, three-phase 
synchronous motor. One transformer provides 110 volts for the 


ruby lamps, and another provides 40 volts for operating the edge- 
printing lamps in series with a trip coil. The main printing lamps 
are operated on 1 15 volts d-c. A simplified switchboard arrangement 
has been developed in which a metal partition separates the a-c. from 
the d-c. circuit (Fig. 7), and a similar partition is provided also 
on the inner side of the board, thus conforming to the most stringent 
underwriters' requirements. Normally, the switchbox cover is 
bolted on. The cover is provided with a door that allows access to 
the fuses and edge-printing lamp switches only. 

Dual Printing. The machine includes provisions for future in- 
stallation of an additional take-up unit on each head for dual 


A simplification of the sensitometric control matte system outlined 
in the previous paper has been effected. Gamma can be determined 
with sufficient accuracy for efficient control by taking readings of any 
two separated positions on the straight-line portion of the H&D curve. 
In order to provide such a check, and to avoid the possibility of using 
the wrong mattes, etc., two slots of the exact widths are cut in a piece 
of opaque stock with a very precise special punch. Corresponding 
slots are cut in a piece of opaque stock spliced into the negative leader. 
This arrangement provides two controlled exposure spots in the 
finished print. The densities are read and entered on a suitable 
form. It is sufficient to know beforehand what the densities should 
be, so that, if any discrepancy is noted, a check can be made to locate 
the source of the trouble. 

This method does not supplant the usual sensitometric tests, but 
avoids the necessity of making the sensitometric strips so frequently. 
At the same time, it avoids also the one- or two-hour delay that now 
occurs between the time a film is developed and the time the final 
sensitometric strip is measured and charted. 

To reduce this delay still further, a special form of flicker com- 
parator, or gamometer, has been developed, which allows the "working 
gamma" to be determined in one or two minutes after the film has 
been developed. To facilitate making a continual check on the 
reduction rate of the developer another instrument called a potenti- 
ometer has been developed. 

The gamometer and potentiometer check each other, as well as 
the printer light setting and the functioning of the developer. They 


allow quick check readings to be made on each roll that is developed, 
and a constant graphical indication of the trend of any variations 
that may occur may be gained by charting the readings. This, in 
turn, admits of quicker action than the present method of reading 
the gamma every two hours or so. These instruments are the 
equivalent of "tolerance gauges" in mechanical work and, of course, 
should be compared with the standard sensitometer and the H&D 
curves at intervals. 


1 HOWELL, A. S., STECHBART, B. E., AND MITCHELL, R. F.: "The Bell & 
Howell Fully Automatic Sound and Picture Production Printer," /. Soc. Mot. 
Pict. Eng., XIX (Oct., 1932), No. 4, p. 305. 


MR. CRABTREE: Is the diameter of the printing sprocket the same as that of 
the model D printer? 

MR. MITCHELL: Yes. The shape of the tooth has been changed slightly to 
provide a little more even slippage characteristic, if I may so express it. The 
diameter is the same, but the arc of contact, that is, the angle of approach 
and release, is different. 

MR. CRABTREE: Is there any reason why this printer will provide better con- 
tact and less slippage than the model D printer? 

MR. MITCHELL: Yes. Apart from the other factors of tooth shape and the 
like, the two films are pressed together by air, as well as by the tension applied 
by the tension rollers. The evenness of travel also helps to eliminate the possi- 
bility of slight slippage. 

MR. CRABTREE : Suppose you flash the film, that is, put a piece of film through 
without a negative, develope it, and then project it; do you notice any variation 
in density? 

MR. MITCHELL: We have made a great number of what are referred to in 
this paper as "fog" tests. Passed in front of the eye, they provide a very sensi- 
tive test. We also placed such a fog test on a sound projector, the idea being that 
if any variations occurred due to any cause whatsoever, they would show up as 
changes of pitch. 

MR. CARVER: Did I understand that the film is pulled through the machine 
by the tension roller, so that the printer sprocket does not pull the film? 

MR. MITCHELL: The printer sprocket pulls the film, but the film is under a 
certain adjustable tension. 

MR. JONES: Are both films under tension in the same direction at one time? 


MR. JONES: How many teeth are in engagement at the aperture, as com- 
pared with the model D printer? 

MR. MITCHELL: In the new printer seven teeth, and in the old printer nine 
teeth exclusive of the one tooth entering and the one tooth leaving the 


E. W. BEGGS** 

Summary. Although the trend in projection lamp development has been toward 
high-intensity, short-life lamps, the need for a lamp that can be operated for long 
periods of time, such as in advertising projectors, has arisen. The design and 
operating characteristics of such a lamp are described, together with the influence of 
airway beacon practice, the problem of existing projectors, projector lens limitations, 
and the effect of voltage change. 

The average home projector is operated only a few hours each 
month. Compared with the investment in the machine and the 
films, the cost of the lamps and the current represents only a relatively 
small part of the annual cost of the outfit. For such equipment, 
therefore, the operator will, in order to project a picture of adequate 
brightness, make use of high-intensity, short-life lamps. Similarly, 
an industrial operator exhibiting a motion picture at a convention or 
a sales meeting, or a commercial projectionist exhibiting pictures 
before a paying audience, may find it desirable and necessary to use the 
25-hour projection lamps. Furthermore, he often chooses to over- 
load the lamps in order to produce the most brilliant picture possible. 

While the trend in projection lamp development in recent years 
has been rightly toward the very maximum in illumination by means 
of high-intensity, short-life lamps, a need has slowly grown for an en- 
tirely different sort. These lamps are needed for projectors that are 
operated long hours, often 8 and occasionally 24 hours each day. 
They are for the most part burned in automatic lantern-slide machines 
operating with a short throw. In machines of this type the requisite 
picture brightness can be obtained rather easily, and the only unfilled 
need seems to be a means for reducing the cost of operation. Such pro- 
jectors are operated in railway stations, restaurants, store windows, 
drug stores, showrooms, and even on building roofs and along high- 
ways. For the most part, they are employed in advertising work. 

* Presented at the Fall, 1933, Meeting at Chicago, 111. 
** Westinghouse Lamp Co., Bloomfield, N. J. 




[J. S. M. P. E. 

In the case of an advertising projector, therefore, a special situ- 
ation is encountered. The size of the audience varies and is often 
small, and the daily operating period is long; so that the success or 
failure of an installation depends largely upon the hourly cost of 
operation. Consequently, the life of the projection lamp is an im- 
portant matter, since that is what largely determines the operating 
cost. The problem, then, is to determine over what period a lamp 
can be operated most economically with little or, if possible, no loss 
of picture brightness. Such a length of time can then be used in 
designing lamps wherein the cost of operation is the most important 

1000 WATT 

200 400 600 800 IOOO 


FIG. 1. Relation between life and operating cost 
of advertising projector lamps. 

consideration. Where the brightness is of greatest importance ob- 
viously short-life, high-intensity lamps will be used, as at present. 


The ideal advertising projector lamp should produce the greatest 
possible screen illumination for the least possible operating cost. 
The items of operating cost in an advertising projector that must be 
considered are the cost of the lamp and the cost of the current. Pro- 
longing the life of the lamp naturally reduces the lamp cost per hour. 
At the same time, however, the lumen output per watt decreases, 
and, consequently, the cost of the current per lumen-hour increases. 
A point is reached where the increase in the cost of the current offsets 


the saving on lamp cost. That point determines the most economical 
operating period for the lamp in question. If such an operating period 
is practicable from the standpoint of the fixture also, it can be estab- 
lished for the service. 

In order to determine the proper operating period for advertising 
projector lamps, the change in operating cost has been calculated for 
a series of lamps. The results of such calculations are plotted in 
the curves of Fig. 1. In computing the data, certain assumptions 
were made regarding the lens system, the wattage capacity of the 
projector, the cost of the lamps, and the cost of current. 

First, it was assumed that the lens system could fully accommodate 
any light source that might be chosen ; that is, that the design of the 
filament of the ideal advertising projector lamp would be suitable if 
the optical parts were designed to conform to it. In extreme cases, 
of course, that would be impossible, since, when the life becomes very 
long, the source becomes too large; but in the series investigated the 
size of the light source was always fairly small, and hence the as- 
sumption was reasonably safe. 

Second, it was assumed that the projector housing could accom- 
modate the higher wattages made necessary by the longer life of the 
lamp. Any cooling of the lamp bulb that might be required should 
be done by natural means in order to minimize the operating cost. 
In most cases the assumption will be justifiable because the range of 
bulb size will not greatly exceed the sizes now in common use, par- 
ticularly when the lamp life is not excessive. 

Third, certain lamp costs must be assumed. For the purpose of 
this analysis, the lamp prices determined are, in most cases, already 
established. Where no prices have been established for the hypo- 
thetical lamps discussed, they were estimated, which means that the 
costs assumed are very nearly correct, although not necessarily exact. 
In the computations, the net lamp cost was assumed on the basis of 
the discount on lamps such as obtains with a small commercial Mazda 
lamp contract. 

Fourth, the cost of electric power varies, depending upon the part 
of the country in which the device is operated and upon the power 
contract according to which the power will be purchased. Assum- 
ing that the device will ordinarily be operated in a small commercial 
establishment, the average cost of power in the United States during 
1932 for that type of installation may be used with safety. That 
figure, according to the reports of the electrical industry, was $0.0409 



[J. S. M. P. E. 

per kw-hr., and in the computations 4 cents per kw-hr. was the figure 

Determined on the basis of the maximum number of lumen-hours per 
dollar, an operating period of 500 hours is most economical for the 
two lamps represented in Fig. 1. The sizes of the source and the bulb, 
for lamps of this size, type, and operating period have been proved 
practicable by five years of service as U. S. airway beacons. Adver- 
tising projector lamps of the wattages cited, therefore, should in 
general be designed for such an operating period, and advertising 
projector apparatus should be designed to accommodate such lamps 
and light sources. Lamps of smaller wattage for advertising pro- 
jector service for the present should be designed for somewhat shorter 













D 50 100 160 

Per Cent of Rated Lamp Life. 

FIG. 2. "Mortality" curve of 500- and 1000-watt lamps, 
indicating the importance of relamping. 

lives so as to maintain the illumination values above the useful 


While many advertising projectors have successfully used airway 
beacon lamps, few if any have fully utilized the facilities made avail- 
able by beacon equipment. It is standard practice in the United 
States to equip each rotating airway beacon projector with an auto- 
matic lamp changer, which introduces a spare lamp into the circuit 
when the active lamp burns out. The lamp changers have functioned 
with little or no trouble for more than five years in more than one 
thousand projectors, and their characteristics are such as could well 
be utilized in advertising lantern slide machines and the like. Such 

Feb., 1934] 



an application would reduce the cost of maintenance and would im- 
prove the continuity of the service. 

At the same time a second feature should be introduced to assure 
the best possible performance of the lamps: the 
lamps should be replaced according to a fixed 
schedule, so timed that, in general, the lamps would 
be renewed before they fail in service. Such a 
process would be based on the "mortality" curve 
of Fig. 2, which shows diagrammatically the sort of 
actuarial data available on Mazda lamps. It 
shows that some lamps fail before and some after 
the average or nominal operating life has been 
attained. The U. S. Department of Commerce 
arranges matters so that in practically every case 
beacon lamps are renewed before the beacon be- 
comes dark. The visits to each station for re- 
lamping are timed so that considerably less than 
1000 hours elapses between visits. Having two 
500-hour lamps in each unit, such service makes 
possible the high degree of dependability now 
achieved in airway lighting. It would seem de- 
sirable to apply similar devices and maintenance 
procedure to advertising projectors for the ad- 
vantages they offer. Projection lamps will func- 
tion very suitably in the lamp changers, and the 
mortality data may be used also to calculate the 
best renewal periods. 

The new bipost base, 1000-watt, T-20, bulb, 
airway beacon lamp is shown in Fig. 3. This 
lamp is also avilable with the mogul screw and mogul prefocus 
type of base. 


Projectors now in use, however, must be provided with economical 
lamps. In many cases they are using standard 50-hour projection 
lamps providing excellent illumination, but with, in some cases, a 
prohibitive operating cost. Those projectors could more profitably 
employ higher-wattage, longer-lived lamps, but the substitution of 
such lamps does not necessarily produce the results desired because 
of optical and, also, mechanical limitations. An analysis of each case 

FIG. 3. The new 
1000-watt beacon 
lamp with the bi- 
post base; auto- 
matic lamp 
changers are avail- 
able for this and also 
the mogul prefocus 



[J. S. M. P. E. 

can be made, requiring certain data to establish definitely the possi- 
bilities offered in each case. 

To choose a suitable lamp for a given projector, the following in- 
formation is required : 

( 1) Size of light source that the lens system will accommodate. 

(2) Filament dimensions of lamps now available. 

(3) Lumen output, prices, and operating life of the lamps. 

(4) Mechanical clearances and cooling requirements of the lamps available. 

(5) Effect of operating voltage on screen illumination, lamp life, and wattage. 

(6) Screen lumen ratios for monoplane and biplane filaments where a change 
of filament structure is to be made. 

FIG. 4. Illustrating the dependence of available illumination upon the size 
of the light source : (A) This condition permits a large increase in the size of 
the source; (B) a good condition for efficient utilization of light, and intense 
screen brightness; (C) lens filled: condition for maximum screen illumina- 

Having all that information, a designer or an operator may deter- 
mine for himself the most suitable lamp for his purpose. Each re- 
quirement will be discussed in some detail below. 


Any stereotype projector now equipped with a projection lamp 
falls into one of three classifications as to lens limitations. Assuming 
that the optical system is arranged to focus the image of the filament 
in the objective lens, the source of light, together with the image 
from the reflector may partly fill, completely fill, or it may over-fill the 
objective lens. If the lens is incompletely filled, changes can be made 
quite freely, whereas, if the lens is full or over-filled, the problem is 
more difficult. Visual inspection of the projector during operation 
will show what are the conditions involved. By looking through a 
dark glass into the objective lens toward the lamp filament, the lens- 
filling conditions can be roughly determined as described below. 

Feb., 1934] 



Fig. 4 shows three lens-filling conditions for monoplane filaments 
and reflected images. A is a condition that is common with 500- 
watt, 50-hour projection lamps when used in lantern slide projectors 
of the standard types now on the market. In this case, a large light 
source can be accommodated, and the illumination of the screen will 
increase approximately in proportion to the light flux emitted by the 
light source until the lens becomes filled. 

When the image of the source is just inscribed in the circle of 
the objective lens as shown in Fig. 4(B), a condition of maximum 


400 f 









95 * 

90 92 94 96 98 100 102 104 106 108 1 10 

FIG. 5. Lamp performance curves, 
in relation to the applied voltage; 
showing light output, power con- 
sumption, and life. 

illumination with little waste is attained. Increasing the size of the 
source will increase the illumination somewhat, but some of the light 
will fail to pass through the lens to the screen. 

In advertising projectors an over-filled lens represents waste in the 
form of initial lamp cost and operating current. In certain instances, 
according to requirements, such a condition may be justified, but 
ordinarily such over-filling should be avoided. 

Fig. 4(C) shows the circle of the lens just circumscribed by the image 
of the filament. This is the condition that provides maximum screen 



[J.S.M. P. E. 

illumination, but it obviously results in appreciable waste : about 22 
per cent of the total flux fails to pass through the objective lens to the 
screen. Increasing the size of the source beyond that shown in Fig. 
4(C) will fail to increase the brightness of the screen. To increase the 
brightness under such a circumstance will require either a brighter 
shorter-lived filament or a filament of biplane construction. 


The filament dimensions, luminous output, recommended operating 
period, list price, and other useful data on various lamps now avail- 
able are given in Table I. When applying these data the following 
general relations among the various types of lamps will be useful in 
computing the results achieved by changes in lamps. 

(1) By doubling the electrical rating of a monoplane filament the projected 
area of the light source will be approximately doubled. 

(2) The output in lumens of a 500-hour lamp is approximately 25 per cent 
less than that of a 50-hour lamp of equal electrical rating. 


Lamps for 100, 105, 110, 115, or 120 Volts 







Fila. Dim. Relative 
Width Height Priceft 



























































Adv. Proj. 







500 f 



















T-20 Short 









T-20 Short 












































* Require forced draft cooling. 
** Preliminary Data. 

t Special lamps. 
ft Subject to change. 

C-13 designates the monoplane, while C-13D designates the biplane filament 
construction. C-138 filaments are monoplane, wider than they are high. 

Feb., 1934] 



(3) The projected area of a biplane filament will be a little more than half 
that of a monoplane filament of equal electrical rating and equal operating 

(4) The projected area of the 500-hour light source will be approximately 
25 per cent greater than that of an otherwise equivalent 50-hour light source. 

Diagram of Optical System 


'''a i 

Screen Illumination Values 


Fila. Hours 


Screen Lumens 
Mirror Mirror 

Operating Cost 
in Cents per Hour 

Current at 44 
Lamp per Kw-hr. Total 








































42,600 f 



3.00ft 8.0 
3.05 8.0 


* Special lamp. 
** Experimental lamp. 

f Preliminary data. 
ft Estimated. 

Image of Filament at the Objective Lens 



50 Hrs. 



500 Hrs. 


50 Hrs. 



500 Hrs. 



500 Hrs. 

500 Hrs. 

FIG, 6, Reference table for various lamps used with a given optical system. 

136 E. W. BEGGS [ J. S. M. P. E, 

The clearances required for the various lamps referred to in Table I 
can be calculated from the bulb designations given. In Mazda 
lamps, the bulb is designated first by a letter that defines the contour 
of the bulb. A T-bulb is a tubular bulb. The number following the 
designating letter defines the maximum diameter of the bulb in Vsths 
of an inch. The maximum diameter of a T-20 bulb, for example, is 
20 /s or 2 x /2 inches. 

The lamps marked with an asterisk require forced ventilation. All 
others listed can be adequately cooled by natural draft. The 300- 
watt T-10 and 750-watt T-20 lamps require only a moderately strong 
draft of cooling air on the bulb. 


In the past, it has been fairly common practice to subject a pro- 
jection lamp to a voltage either greater or less than its normal rating. 
This effects lamp performance as shown in the curves of Fig. '5, from 
which can be calculated the expected average life, luminous output, 
and wattage of any lamp for which the data are required. 


Substitution of Long Life Lamp. A practical example of the 
application of the principles discussed here would be to select a lamp 
suitable for replacing a 500-watt, 50-hour projection lamp in a typical 
lantern-slide projector used for advertising. In this case, the extent 
to which the lens is filled will be assumed to be as shown in Fig. 4 (A). 
It is apparent that the size of the source can be increased without waste. 

Assuming that it is desirable to maintain or improve the illumina- 
tion, the 1000-watt beacon lamp will be selected, and its operating 
costs and illumination data will be determined. 

Fig. 4(B) illustrates approximately the condition .attained by using 
such a source. As it is fully accommodated by the particular ob- 
jective lens that is used, the brightness of the screen will be increased 
in proportion to the luminous output of the two lamps. The increase 
of screen illumination can be calculated by means of the following 
formula, where I\ m is the new value and Im the original value 
for screen lumens : 


/1000 = /500 X = 156% Of 7 6 00 


This shows that theoretically an increase in illumination can be 
achieved under the conditions assumed. A series of tests conducted 


on a typical commercial projector afforded the data on illumination 
and operating cost given in Fig. 6. 

In these tests the objective lens was filled with light to the extent 
shown in the photographs. The over-filling of the lens by the 1000- 
watt beacon lamp produced an illumination somewhat less than the 
theoretical value calculated by the formula given above. The data 
are exact enough, however, to confirm the theory of the formula. 

Use of Biplane Filament. The use of the biplane filament would 
be appropriate where the screen illumination is to be increased over 
what is possible with the monoplane filament. The data in Fig. 6 
show the effectiveness of such a lamp of 2000 watts in a lens system 
for which the 2000-watt monoplane source is altogether too large 
and, therefore, relatively ineffective. 

Effect of Undervoltage. To show the effect of undervoltage an ex- 
ample can be computed from the data given in Fig. 5. Assuming 
that it is desired to double the life of the lamp, and that a reduction 
in illumination will not seriously affect the results, the final operating 
conditions can be computed from the curves which indicate a life 
of 200 per cent of the normal at a voltage of 95 per cent of the rated 
value. Let /wo represent the illumination when the lamp is operat- 
ing at the rated voltage, or 100 per cent. Let 7 95 be the illumination 
when the lamp is operated at 95 per cent of its normal voltage, which 
would be the case if a 11 5- volt lamp were operated at 109.3 volts. 
The resulting illumination will be: 

/95 = /ioo X 0.83 = 83% of 7ioo 

Therefore, in this instance, the illumination would be reduced 
17 per cent. Similarly, the wattage will be reduced from 100 to 92 J /2 
per cent, and the life increased from 100 to slightly over 200 per cent. 
Assuming a 1000-watt T-20 lamp, the operating cost would therefore 
be reduced approximately 16 per cent, with a reduction in picture 
brightness of 17 per cent. 


High-intensity, short-life incandescent light sources are needed for 
projecting pictures before audiences who demand the maximum 
screen brilliance. The lamps that have been developed during the 
past year or two have filled that need, which was most urgent for motion 
picture work. Lantern-slide machines, particularly of the automatic 
type with which the requisite screen brightness is relatively easy to 

138 E. W. BEGGS 

attain, and when the lamps burn long periods each day, present a 
radically different problem. Here the application of lamps of higher 
wattages with filaments designed for longer lives seems to be the 

Higher wattages and larger light sources and bulbs will present new 
problems as regards lens and fixture design. In general, monoplane 
filaments will be applied for economy; but on occasion, when the very 
maximum of illumination is needed, the biplane type of source may 
be used. Lamp changers of the airway beacon type, together with 
periodic lamp replacement schedules, are recommended for reducing 
the maintenance costs and to increase the continuity of service to a 

J. O. BAKER** 

Summary. Improvements that have been made in the production and reproduction 
of sound on 16-mm. film are briefly described, together -with some of the problems 
that have been encountered. The two methods of obtaining sound-on-film, namely, 
recording and optical reduction printing, are discussed. It is now possible to reduce 
optically frequencies as high as 9000 cycles on 16-mm. film. Using a practical design 
of 16-mm. reproducing equipment it is possible to obtain sound from the 16-mm. 
film of as good quality as the sound that was formerly obtained from the average 35-mm. 
theater equipment. 

The purpose of this paper is to describe briefly the improvements 
that have been made in the production and reproduction of sound on 
16-mm. film and some of the problems that have been encountered. 
The task of overcoming those problems has led not only to methods 
of producing high frequencies on 16-mm. film, but incidentally has 
also assisted in the production of better 35-mm. sound. A comparison 
of the two and their relative frequency ranges will serve to give an 
idea of the improvements that have been made and the quality of 
reproduction that can be obtained from 16-mm. film. 

In the past it has been the commercial practice to limit the high- 
frequency response on 16-mm. film to approximately 4000 cycles, 
and on 35-mm. film to approximately 6000 cycles. General im- 
provements that have been made in 35-mm. recording, processing, and 
reproducing have extended the frequency range to 9000 and 10,000 
cycles. Similar improvements in the 16-mm. processing have ex- 
tended the 16-mm. frequency range to 6000 cycles without serious 
attenuation, while frequencies as high as 9000 cycles can be placed 
on the film and reproduced on high-fidelity reproducing equipment. 
As pointed out in a previous paper 1 there are two methods of ob- 
taining sound on 16-mm. film from 35-mm. recordings: namely, 
by re-recording and by optical reduction printing. 

Re-recording has the advantage that the frequency characteristic 

* Presented at the FalU 1933, Meeting of Chicago, 111. 
** RCA Victor Co., Camden, N. J. 




[J. S. M. P. E. 

FIG. 1. Comparison of frequency characteristics, using S. M. P. E. stand- 
ard sound test film : (A ) measured galvanometer deflection ; 35-mm. recording ; 
(B) measured film output; 35-mm., 0.5-mil reproducer slit; (C) measured film 
output; 16-mm., 0.5-mil reproducer slit. 

35-mm. negative 

16-mm. optical reduction print 

FIG. 2. 4000-cycle sound tracks: 35-mm. negative and 16-mm. 
optical reduction print. 

Feb., 1934] 



can be altered by compensation if found necessary or desirable. 
However, it has a disadvantage in that losses are introduced, which 
are somewhat difficult, but not impossible, to overcome. These 
losses are due to two causes : irradiation within the emulsion, and the 
finite width of the recording slit. 

Optical reduction of the sound track has two distinct advantages. 
One is that no additional slit losses are introduced. The other is 
the reduction of irradiation losses, which is due to the specular form 

35-mm. negative 

16-mm. optical reduction print 

FIG. 3. 6000-cycle sound tracks: 35-mm. negative and 16-mm. 
optical reduction print. 

of the printing light. In other words, the printing light enters the 
film more nearly parallel than in the case of contact printing where 
diffused transmission of light is utilized. 

The curves of Fig. 1 are intended to show a comparison between the 
frequency characteristics of the 35-mm. and the 16-mm. reproduced 
sound. These curves were made with the Standard S. M. P. E. Sound 
Test Film. Curve A is the measured galvanometer deflection of the 
35-mm. recorder having a V2-inil recording slit. From this curve 
it can be seen that the compensation required in the recording to 



[J. S. M. P. E. 

produce a flat frequency response when the film is reproduced in a 
no-loss reproducer is 4.7 db. at 6000 cycles and 9.5 db. at 10,000 

Curve B is the measured output from the 35-mm. negative, using 
a 1 /2-mil reproducer scanning slit. If this curve is corrected for the 
reproducer slit loss the characteristic will be substantially flat. 

Curve C is the measured output from a 16-mm. print, optically 
reduced from the 35-mm. negative and reproduced on a reproducer 
having a 1 /2-mil scanning slit. It should be noticed that the film and 
reproducer loss is 11 db. at 6000 cycles and 20 db. at 9000 cycles. 

35-mm. negative 

16-mm. optical reduction print 

FIG. 4. 9000-cycle sound tracks: 35-mm. negative and 16-mm. 
optical reduction print. 

Since the loss due to the 16-mm. reproducer scanning slit is approxi- 
mately 3 db. at 6000 cycles and 6 db. at 9000 cycles, the film loss is 
reduced to only 8 db. at 6000 cycles and 14 db. at 9000 cycles. 

Figs. 2, 3, and 4 are photographs of the 4000-cycle, 6000-cycle, and 
9000-cycle sound tracks, respectively, of both the 35-mm. negative 
and the 16-mm. optical reduction positive films. The filUng-in of 
the valleys of the 16-mm. sound tracks is due to irradiation of the 
16-mm. emulsion. These photographs were magnified 28.5 to 1. 

Feb., 1934] SlXTEEN-MM. SOUND-ON-FILM 143 

By this method the sound printed on the 16-mm. positive will be 
free from distortion, except to the extent that it may be present 
in the 35-mm. negative and introduced due to improper processing 
of the film. It is possible, therefore, to produce a high-fidelity 16-mm. 
sound track which may be reproduced on suitable reproducing 
equipment. The 16-mm. reproductions that have been made demon- 
strate the practicability and uniformity with which acceptable 
16-mm. sound can be obtained from film. 


1 BATSEL, C. N., AND BAKER, J. O. : "Sound Recording and Reproducing 
Using 16-Mm. Film," J. Soc. Mot. Pict. Eng., XXI (Aug., 1933), No. 2, p. 161. 


Summary. The economics and practical application of color cinematography, 
its problems in production, and its laboratory requirements in the industrial field 
are briefly reviewed. 

Many users of the motion picture in the advertising and the 
selling fields find a definite need for color in their productions. 
Several years ago, it was realized that the sale of business films 
might be stimulated if some practicable and economical color process 
were available for commercial use. 

In surveying the possibilities of color photography in the industrial 
field it was necessary to consider the following points due to vast 
differences in production set-up between theatrical and non-theatrical 
production : 

(1) The design and mobility of cameras and equipment. 

(2) Developing and printing facilities. 

(3) Print costs. 

(4) Projection equipment. 

After a thorough review of the field and after many tests were 
made, it was decided that 16-mm. color films, such -as were available, 
could not be considered practicable in this field because of the fol- 
lowing limitations : 

(1) Restriction in duplicate prints. 

(2) Size of picture. 

(3) Limitation to non-theatrical field. 

It was found also that the Technicolor process, using Technicolor 
cameras, prints, and service, would not fall within the requisite non- 
theatrical production costs and the necessity for prompt print de- 
livery in small quantities. 

It was decided, therefore, to adopt the bi-pack process for making 

* Presented at the Fall, 1933, Meeting at Chicago, 111. 
** Ray-Bell Films, Inc., St. Paul, Minn. 


industrial films in color. Through the assistance of W. T. Crespinel 
and his laboratory (then Multicolor) this company produced its 
first industrial color film aboard a transcontinental train in May, 
1930. Since that date some all-color releases and some part-color 
releases have been produced with considerable success and satis- 
faction. As long as laboratories equipped to process color prints 
are able to deliver a good standard of prints it is quite probable that 
color will continue to be used in industrial films, and that an increase 
in color footage in that field will be seen before long. 

Why the bi-pack process has proved to be the most adaptable 
to the industrial field will be understood from the following account. 

Camera Equipment. The regular black-and-white Bell & Howell 
or Mitchell cameras with certain minor changes in the pressure ad- 
justment of the aperture plate may be used in a bi-pack process, 
therefore making unnecessary any great outlay of money for new 
camera equipment. Magazines for the double bi-pack negatives 
are not expensive, and do not add any particularly cumbersome 
apparatus to the cinematographer's equipment. 

Developing and Printing Facilities. After exposure of bi-pack 
negatives it is possible to develop both the panchromatic and the 
red-dyed or orthochromatic negatives in a manner very similar to 
that for ordinary black-and-white negatives. It is quite necessary, 
of course, to make sufficient tests in order that a proper balance be- 
tween the two negatives is maintained in order to achieve a correct 
balance of color in the print. No added equipment is necessary 
for this process other than a tank of hydrosulfite bleach for clearing 
the dye from the front negative. 

Print Costs. Industrial film users buy relatively few prints, but 
always insist on buying most economically; it was found that both 
nitrate and safety stock color prints could be furnished at about 
twice the cost of black-and-white prints. Good service from the 
laboratories in Hollywood has been maintained, and prints can be 
delivered within twelve days after the negative has been shipped from 
the laboratory. 

Projection Equipment. Many industrial film users have 35-mm. 
projectors, and in using the bi-pack process no change of any kind 
is necessary in the projection equipment. It is also possible to use 
the films in theaters. 

There are numerous other advantages of such a color process. It is 
impossible to use large crews in shooting industrial films, and a camera- 

146 R. H. RAY AND H. W. CRESS [J. S. M. P. E. 

man, assistant cameraman, and director can easily handle a silent 
scenic job in color, using this system. It is often necessary to pack 
the equipment and travel over mountain trails, often to places 
where the motion picture camera has never been before and, under 
such circumstances the less equipment and personnel needed, the 
easier it is to obtain such contracts. The cameramen, when shoot- 
ing color pictures, have followed a few rules that have aided them in 
obtaining good color results with the bi-pack process during the past 
several years. These rules will, no doubt, be helpful to those contem- 
plating bi-pack color photography. 

In making scenic exterior shots it is quite necessary that the at- 
mosphere be clear and the sunlight brilliant. It is a good plan to 
nick the film at the scene change and make careful tests when there 
is any decided change in the type of scenes made on the same roll 
of negative. Care should be taken with relatively short ends left 
in magazines for any length of time as this will tend toward "breath- 
ing" trouble. In making interiors particular care should be taken 
to light up the floor, especially if a dark rug or dark upholstered 
furniture is used on the set. Finally, the cameraman should be 
particularly careful as to the particular colors that are attempted. 
It is well to adhere to the reds, orange, blue-greens, blues, and browns. 
With these thoughts in mind good results should be obtained and 
every one satisfied with the production. 


MR. CRABTREE: You said something about the reproduction of yellows in 
the prints made by Cinecolor. Can you tell us more about it? 

MR. RAY: It is very new. We have been trying some experiments with titles 
using the Eastman background negative. Two niters were recommended, but 
I haven't the data with me. 

MR. CRABTREE: Is it the final print that shows the yellows, as well as the 
reds and blues and greens? How are the blues? 

MR. RAY: They are blue-greens. 

MR. CRABTREE: And the reds? 

MR. RAY: The reds are orange. I am sure you know the limitations of two- 
color processes. 

MR. CRABTREE: Yes, but I was just wondering how you get the yellows. 

MR. RAY: Yellow is an over-all dye that is put on the film. 

MR. CRABTREE: Probably imbibed on the film? 

MR. RAY: Yes; and in clear whites it will come through yellow. 

MR. CRABTREE: But how do you determine where to put the yellow? Is 
it used only in connection with titles? 

MR. RAY: Titles and cartoons, 

Feb., 1934] COLOR FOR FlLMS 147 

MR. FARNHAM: What form of lighting equipment was used for the interior 

MR. RAY: Incandescent. 

MR. CRABTREE: Perhaps Mr. Farnham is wondering whether you used the 
overvolted lamps. 

MR. RAY: No; we haven't tried those yet, but we hope to do so very soon. 
That should improve the color efficiency considerably. 

MR. CRABTREE: It is of historical interest that Dr. P. G. Nutting and Mr. 
J. G. Capstan 7 used overvolted tungsten lamps for color photography in the East- 
man Research Laboratories in 1914. A small stage about 15 feet square was 
equipped with special 1000-watt lamps supplied by the General Electric Co. 
and, of course, they were considerably overvolted. In front of the lamps window- 
glass with chicken netting embedded in it was fitted so as to protect the actors 
in case the lamps should burst. A stream of cold air was blown over the lamps 
to keep them cool. The life of the lamps was relatively short, but that of the 
modern lamps is surprisingly long. 

How much more can you charge for such colored pictures than you can for 
black-and-white ? 

MR. RAY: So far we have found it necessary to increase our costs about 
50 per cent for color over black-and-white. 

MR. CRABTREE : You certainly can't make much profit at that rate when you 
have to use twice the footage of negative. 

MR. RAY: In the additional prints that are used there is a little more 
profit in the sale of color prints than in black-and-white. 

MR. CRABTREE: Why is it that people are willing to pay for the color in 
this class of work, and yet the producer of photoplays is not interested? 

MR. RAY: Color is particularly adaptable to selling things in which color 
is dominant. It provides an additional sales advantage. 

MR. CRABTREE: But the producer is often interested in selling sex appeal, 
and I think color would add to that. 

MR. RAY: That is the point. In the industrial field color isn't always ap- 
plicable to every subject, but when it can be used advantageously we are able 
to sell our customer on the idea of the additional expense of color over black- 

MR. MURRAY: Has there been any demand for 16-mm. color in the industrial 

MR. RAY: Yes. I had a sample the other day from Hollywood, made by 
a 16-mm, two-color process. 



Summary. Some of the more common acoustical defects of theaters, suck as the 
transmission of extraneous noises into the auditorium, the emanation of sounds from 
vibrating or rotating machinery, the reflection effects of curved surfaces, and reverbera- 
tion, are discussed. The manner of treating auditoriums for obviating such effects 
is described briefly. 

In presenting this paper the authors have borne in mind numerous 
theaters that are constructed and that have, or have not, been 
treated acoustically for the purpose of affording the critical patron 
the degree of satisfaction to which he is entitled. The average 
theater owner, manager, or projectionist knows very little about 
how the reproduction sounds in the house; just as long as the projection 
room monitor is loud enough and the projectionist can hear dis- 
tinctly, everybody is satisfied. 

The degree of satisfaction that the patron derives from a mo- 
tion picture theatre presentation depends on many factors, viz., 
the acoustic properties of the house, the projection, the sound equip- 
ment, heating and ventilation, seating comfort, etc. Most theater 
managers are acquainted with the projection, ventilation, seating, 
etc.; yet, in spite of all the money that is involved in such factors in 
the class A houses, and the high degree of perfection attained in 
those directions, the manager does not always find that his patrons 
are satisfied to the extent he would like them to be. It often happens 
that many of the audience have to strain their ears to hear the sound, 
or are wondering when the level of the sound is going to be reduced 
to a more comfortable volume. Again others may not be able to dis- 
tinguish what is being said. 

Cases have come to our attention wherein managers and owners 
have spent large sums in improving the sound equipment, only still 

*Presented at the Fall, 1933, Meeting at Chicago, 111. 
**United States Gypsum Co., Chicago, 111. 



to be confronted with defective sound. Had the managers given 
their problem a little thought, and had their knowledge of the acous- 
tics of the house been enlarged, those sums might not have been 
spent. Poor quality of sound will tend to create dissatisfaction 
much more rapidly than any other kind of defect. It is trying on 
one's nerves not to be able to understand all that is being said. 

It is the purpose of this paper to acquaint the theater manager and 
the projectionist with some practical knowledge of where to look for 
acoustical defects and how to remedy them. When planning to 
build a theater, it seems logical that the owner should engage a 
reputable acoustical engineer to check the design of the intended 
auditorium so that it will be acoustically correct. The location, 
size, and shape of the interior surfaces, the placement of vibrating 
machinery, the ventilating system, and numerous other factors that 
will be shown to be important, should be considered very carefully 
and thoughtfully. Once a house has been built, acoustical correction 
is very often more costly than it would be if the acoustical factors 
were taken into consideration in the original design. Very often 
the house that has been corrected is not as good acoustically 
as it might have been had it been built properly, from the acoustical 
point of view. 

There are four major points to be considered: first, extraneous 
noise. Motion picture theaters are always built where the largest 
possible number of persons pass their doors; on streets where there 
is adequate transportation; in most cases at the crossing of busy 
thoroughfares. Acoustically, such locations are not the best. But 
since potential box-office receipts always govern the location, outside 
noises must be excluded by proper sound insulation if the house is to 
be profitable. 

Noises coming through the front doors are much easier to exclude 
than those coming through the ground, up through the floor, and into 
the house. Noises coming through the front foyer doors may be 
reduced almost to inaudibility by means of entrance doors on the 
street end of the lobby. If that does not suffice, and additional 
reduction is needed, the ceiling and part of the side walls of the lobby 
may be covered with absorbent material. The foyer should be 
heavily carpeted, and thick plush drapes should be used for decorat- 
ing it. 

When a house has fife doors that open to a noisy street or alley, 
with small vestibules leading to them, the vestibules should be laid 

150 G. W. BAKER AND M. A. SMITH [j. S. M. P. E. 

with padded carpet, and absorbing material should be added to the 
ceilings and walls. Drapes should be hung over the opening between 
the house and the vestibules. 

The ducts of the ventilating systems should be lined with absorbent 
material. There are sectional absorbents on the market that can 
be conveniently adapted to almost any kind of ventilating duct. 
Windows should be made with double or triple panes, spaced apart 
and floated in felt. 

Noises from the ground coming through the floor usually require 
considerable, expensive treatment, and should be studied by a cap- 
able engineer. In most cases the noise may be reduced to inaudibility 
by placing a floor mounted on a resilient structure over the old floor. 
vSuch resilient constructions absorb the vibrations, and prevent 
them from entering the house. Noises that are transmitted through 
the house walls may be reduced by placing resiliently mounted faces 
on the walls, for the purpose of damping the vibrations. 

The second major factor to be considered is the noise emanating 
from unbalanced or vibrating machinery. When a machine is 
placed on a floor or wall the noise that it creates is amplified because 
of the larger vibrating surface. The larger the area the louder the 
noise. Projection arc generators are frequent causes of such trouble. 
When they are located on the projection room floor, the sound is 
transmitted by the floor into the house. When located in the base- 
ment the noise travels upward through the columns that support 
the floor, and thence into the auditorium. Usually the projectionist 
will place a pad beneath the generator, expecting the pad to eliminate 
the noise. Sometimes such a plan works, but usually the loading 
of the pad is incorrect, and the low-pitched noise still remains 

When placing resilient pads beneath generators, motors, pumps, 
blowers, etc., correct loading and flexibility may be attained by trial 
and error; but much money will be saved by seeking advice from 
reputable concerns that manufacture such resilient bases. Where 
there are pipe connections to a machine mounted on a resilient base, 
an effort should be made to connect the pipes resiliently, because 
otherwise they may transmit the offending vibrations to the walls 
and into the house. 

Exhaust fans placed in the walls leading to the outside are constant 
sources of noise. In the summer months most of the fans are useless, 
because they are too noisy, and mask the sound. By properly'de- 


signing the fan blades and guards, and properly balancing the 
rotating parts, the vibrations, and hence the noise, may be reduced; 
but generally, with all the care that is taken, there is sufficient 
vibration to be objectionable. If the fan is placed in an opening 
in a wall the noise is amplified. Resilient constructions placed 
between the fan mounts and the wall will remedy the defect. Here 
again proper proportioning of loading and flexibility must be given 

The third factor to be considered is one in which surgical treat- 
ment is usually necessary. Objectionable reflecting surfaces, which 
have been discussed extensively in the literature, are present to a more 
or less degree in every house. 

The worst kind of surface is the curved surface, which acts toward 
sound just as a curved mirror acts toward light : If convex, the sound 
is dispersed; if concave, it is concentrated to a focal point. In the 
theater there are curved ceilings, curved balconies, curved rear walls, 
domes, and barrel-shaped surfaces. Curved ceilings with the center 
of curvature at the floor line, in the audience, or slightly above, 
present a very difficult problem. When such a condition exists the 
auditors at the focal point of the reflecting surface hear the sound at 
a very high level of volume. Those outside the focal area will hear 
less than the normal volume. Such a defect should have been 
avoided in the original design. If such a troublesome surface exists 
in a house already built, it may be partly corrected by deep coffering. 
Sometimes it is advantageous to line the surface with an effective 
absorbent. A new ceiling of proper shape, constructed below the 
offending surface, will remedy the defect. Curved surfaces that have 
their centers of curvature below the floor line or high above the 
audience generally present little difficulty. 

Curved rear walls and balconies offer the same objections, and are 
dealt with in the same manner. Objectionable concentration of 
sound by curved panels may be obviated by breaking up the panel into 
divergent reflecting surfaces. 

One way of determining whether there are any troublesome sur- 
faces is to walk slowly about the auditorium while listening to a per- 
son on the stage speaking in a normal tone of voice. If there are any 
areas where the sound is unusually loud they should be noted. 
By standing in such areas one can usually determine where the 
offending surfaces are located by noting the general direction of the 
sound. The observer can not determine the location of the reflecting 


surfaces by clapping his hands, as the source of the sounds should be 
located on the stage, and not in the audience. 

Sometimes echoes will be heard; the auditor will hear the sound 
coming directly from the stage, and an instant later will hear the same 
sound reflected from some surface. Echoes can be detected only 
when the difference between the length of the direct path and that of 
the reflected path is eighty feet or more. 

The fourth major factor to be considered is that of reverberation. 
This might have been mentioned first, since many engineers regard 
it to be the only factor of importance. However, the acoustical 
qualities of an auditorium depend upon the others as well. 

Reverberation is the continuance of a sound for such a length of 
time that it interferes with a sound made subsequently. The longer 
it takes for sound to become reduced to inaudibility, the more diffi- 
cult it is to hear clearly. The time of reverberation (the time re- 
quired for a sound of a definite intensity to become inaudible) can 
be calculated and measured. Knowing the time, one can install 
absorbing material over a sufficient area to reduce it to the proper 
limit. Excessive reverberation is bad in any case. 

After the proper amount of absorption required for acoustical cor- 
rection has been determined, the choice of the surfaces to be so treated 
should be made. The usual practice is to place the material on the 
ceiling of the house proper, under the balcony, on the rear wall, 
and the rear portion of the side walls. The best practice is to dis- 
tribute over almost the entire surface of the walls and ceiling a 
material of only fair absorbing power. 

When material of great absorbing power is placed under the bal- 
cony, the hearing becomes poor in the rear seats as the house fills 
with people. The absorption of the audience, plus that of the 
treated balcony ceiling, causes the sound level to be lower under the 
balcony than in the main body of the house. Patrons sitting be- 
neath the balcony complain, and the usher causes the sound level 
to be raised. Those under the balcony are satisfied, but those in 
the first few rows near the stage complain that the sound is too loud. 

The stage and the front portion of the side walls should not be 
covered with absorbing material, and the absorbing power of 
the auditorium should be maintained as constant as possible. The 
use of seats, the absorbing power of which is about equal to that 
of an auditor, is advisable. The acoustical absorption of the patron's 
clothing is substituted for that of the seat he occupies. 

Society Announcements 


The regular monthly meeting of the New York Section was held on January 
10th, at the Eastern Service Studios, Inc., Long Island City, N. Y. Mr. T. Keith 
Glennan, Vice-President and General Manager, and Mr. R. O. Strock, Chief 
Sound Engineer, delivered talks on "The Studio and Its Operation." After the 
talks, the main stage, the lower floor, the power and recording rooms, and the 
carpenter shop were opened for inspection. Examples of the work done by the 
studio were projected for the entertainment of the members, including a new 
feature picture, His Double Life. 

Through the courtesy of Dr. Harvey Fletcher and the Bell Telephone Lab- 
oratories a demonstration of the "Transmission and Reproduction of Speech 
and Music in Auditory Perspective" was held in the auditorium of the Engineer- 
ing Societies Building at New York, N. Y., on January 30th, to which the members 
of the New York Section were invited. Full details of the demonstration 
will be given in the March issue of the JOURNAL. 


The regular monthly meeting of the Chicago Section was held on January llth, 
at the Electrical Association, Chicago, 111. The subject of the presentation was 
"Microphone Technic in Recording," a discussion of the practical applications 
of microphones to motion picture film recording, supplemented by an exhibition 
of microphones of various types, such as condenser, crystal, moving-coil, and 
velocity. Messrs. J. E. Jenkins, of Jenkins & Adair, Inc., and W. Hotz, of Burton 
Holmes Lectures, Inc., assisted in the presentations, and microphone concen- 
trators were exhibited through the courtesy of the National Broadcasting Com- 


The regular monthly meeting of the Projection Practice Committee was held 
on January 24th, at the Paramount Building, New York, N. Y., at which time 
further progress was made in the preparation of the report of the Committee to 
be presented at the Spring Convention. The major portion of the evening was 
devoted to a discussion of the a-c. carbon arc, which the Committee is investi- 
gating from the standpoint of practical projection. 


Coming: A Fourth Tear of 


OINCE Eastman Super-sensitive Panchro- 
^ matic Negative was introduced early in 
1931, its revolutionary qualities have fulfilled 
every hope and prediction of its sponsors. It 
has helped cameramen and producers so tre- 
mendously ... it has affected the motion pic- 
ture art so profoundly ... it has contributed to 
so many cinematic triumphs, that a further 
prediction can now be made: In its fourth 
year, as heretofore, this Eastman film will 
be an important factor in the most conspicu- 
ous motion picture achievements. Eastman 
Kodak Company. (J. E. Brulatour, Inc., Dis- 
tributors, New York, Chicago, Hollywood.) 

EASTMAN Super-sensitive 

Panchromatic Negative 




Volume XXII MARCH, 1934 Number 3 



Equipment for Recording and Reproducing Sound with Photo- 
Film A. F. CHORINE 157 

Standard S. M. P. E. Visual and Sound Test Reels 173 

High-Fidelity Lateral-Cut Disk Records F. C. BARTON 179 

Open Forum: Should Studio Recording Equipment Compen- 
sate for Theater Reproducing Characteristics ? 183 

An Automatic Change-Over Device A. PRITCHARD 186 

The Control Frequency Principle 


The Rotambulator A New Motion Picture Camera Stand. . . 

J. A. DUBRAY 200 

Book Reviews 206 

Society Announcements 208 





Board of Editors 
J. I. CRABTREE, Chairman 


Subscription to non-members ,$8.00 per annum; to members, $5.00 per annum, 
included in their annual membership dues; single copies, $1.00. A discount 
on subscriptions or single copies of 15 per cent is allowed to accredited agencies. 
Order from the Society of Motion Picture Engineers, Inc., 20th and Northampton 
Sts., Easton, Pa., or 33 W. 42nd St., New York, N. Y. 

Published monthly at Easton, Pa., by the Society of Motion Picture Engineers. 

Publication Office, 20th & Northampton Sts., Easton, Pa. 
General and Editorial Office, 33 West 42nd St., New York, N. Y. 
Entered as second class matter January 15, 1930, at the Post Office at Easton, 
Pa., under the Act of March 3, 1879. Copyrighted, 1934, by the Society of 
Motion Picture Engineers, Inc. 

Papers appearing in this Journal may be reprinted, abstracted, or abridged 
provided credit is given to the Journal of the Society of Motion Picture Engineers 
and to the author, or authors, of the papers in question. The Society is not re- 
sponsible for statements made by authors. 

Officers of the Society 

President: ALFRED N. GOLDSMITH, 444 Madison Ave., New York, N. 
Past President: JOHN I. CRABTREE, Kodak Park, Rochester, N. Y. 
Vice-President: WILLIAM C. KUNZMANN, Box 400, Cleveland, Ohio. 
Vice-President: OMER M. GLUNT, 463 West St., New York, N. Y. 
Secretary: JOHN H. KURLANDER, 2 Clearfield Ave., Bloomfield, N. J. 
Treasurer: TIMOTHY E. SHEA, 463 West St., New York, N. Y. 


EUGENE COUR, 1029 S. Wabash Ave., Chicago, 111. 
HERFORD T. COWLING, 7510 N. Ashland Ave., Chicago, 111. 
RALPH E. FARNHAM, Nela Park, Cleveland, Ohio. 
HERBERT GRIFFIN, 90 Gold St., New York, N. Y. 
EMERY HUSE, 6706 Santa Monica Blvd., Hollywood, Calif. 
WILBUR B. RAYTON, 635 St. Paul St., Rochester, N. Y. 
HOMER G. TASKER, 41-39 38th St., Long Island City, N. Y. 



Summary. The work of the Central Laboratory of the All- Union Electrical 
Trust, Leningrad, U. S. S.R., under the direction of the author, in connection with 
the study, design, and manufacture of motion picture recording and reproducing ap- 
paratus, is described. The present installment of the article deals with the research 
on light modulators, variable width and variable density; ordinary and noiseless 
recording systems; various forms of recording light sources; and the relative char- 
acteristics of the different kinds of modulator, viz., Western Electric, RCA, the 
author's, Kerr's cell, and the flashing lamp. The article will be concluded in the 
April, 1934, issue of the Journal. 

In 1926 the study of problems of the sound picture was begun in 
this laboratory, and since that time work has been conducted both 
on the theoretical aspects of the problems and on the design and 
construction of equipment for recording and reproducing sound. 
At the outset, light modulators for film recording constituted the 
chief problem. Their design was subject to such wide controversy 
that it was necessary to undertake extensive experiments with 
existing modulators in order to determine their relative excellence. 
Indeed, on recording and reproducing systems in general, much of 
the data previously accumulated seemed to be of questionable 
accuracy; and since it was impossible to acquire elsewhere the 
equipment with which to check the data, we had to develop our own 
methods and equipment for the purpose. Some of the methods were 
not only original but were devised to deal with problems of major 
importance, and it is such methods, and the equipment designed by 
us for practical use, that are described in this paper. 


The first light modulator that was developed is shown schematically 
in Fig. 1 and more concretely from the top and side in Fig. 2. A 

* Received October 2, 1933. Limitations of space have necessitated abbrevia- 
tion of the original paper. 

** Director, Central Laboratory, All-Union Electrical Trust, Leningrad, 
U. S. S. R. 




[J. S. M. P. E. 

bronze or aluminum ribbon is fixed in the field of a magnet, so that 
when sound currents pass through the ribbon, it vibrates perpen- 
dicularly to the lines of magnetic flux. An optical system, consisting 
of a light source A, objectives B and C, and a cylindrical lens D, 
records these vibrations on the film F. The first objective focuses an 
image of the slit MI (Fig. 2) in the plane of the ribbon. The second 
objective produces an image of the ribbon on the film, and the 
cylindrical lens (OB) produces an image of the slit M 2 reduced to 0.02 



FIG. 1. (Upper) Schematic arrangements of first light 
modulator developed in the author's laboratory. 

FIG. 2. (Lower) Details of modulator illustrated in 
Fig. 1. 

mm. Thus there falls upon the film a thin strip of light whose 
length is determined by the position of the ribbon. When the ribbon 
vibrates and the film moves, an oscillogram of the sound-current 
actuating the ribbon is produced. In order to darken half the track a 
ribbon 0.25 mm. wide would be necessary, and since the ribbon is 
only 0.17 mm. wide, a small plate 0.30 mm. wide is attached to it. 

By simply turning the ribbon and the magnet ninety degrees, so 
that the ribbon is parallel to the axis of the cylindrical lens, the sys- 
tem can be changed from a variable width to a variable density 

Mar., 1934] 



recording system. The vibration of the ribbon when so placed 
varies the amount of light falling upon the film over the whole 
width of the track. The latter use of the system offers a number of 
advantages, due to the fact that the maximum displacement of the 
ribbon needs be only 0.05 mm., instead of the 0.12 mm. required for 
variable width recording. Hence the small plate may be discarded, 
and weaker currents can be employed, tending less to change the 
vibrational characteristics of the ribbon due to heating. 

The first model was completed by the author in 1927, with the 
assistance of his close associates, Litvinsky and Smirenin, and the 
senior laboratory worker, Padkovsky. The modulator is shown in 
Fig. 3, in position to be tested by the photoelectric cell shown in the 

FIG. 3. First model of light modulator, in position to 
be tested by the photoelectric cell in the foreground. 

foreground. It was soon apparent that improvements in details 
could be effected in this simple form of apparatus. In the first 
place the frequency characteristic showed a peak ascribable to the 
tendency of the ribbon to vibrate at its natural frequency. Second, 
it appeared that many advantages might be gained by further mag- 
nifying in some way the effects of the vibrations of the ribbon on the 
amount of light falling on the film. 

Theory indicated that the frequency characteristic would be im- 
proved by making the natural frequency of the ribbon as high as 
possible and by damping its vibration. For a given length of ribbon, 
the former can be accomplished by decreasing the mass and increasing 
the tension as much as possible, and to this end duralumin was used. 
Experience showed that immersing the ribbon in petroleum oil of 

160 A. F. CHORINE [j. s. M. P. E. 

carefully chosen properties was a satisfactory method of damping 
its vibrations. So damped, the response of even a bronze ribbon 
with an attached plate varied not more than 3 db. at frequencies up to 
7(XO cycles per second. 

Experiments were also conducted in which the ribbon was left 
undamped and an electrical filter was used to compensate for the 
resonance of the ribbon. That procedure had the disadvantages of 
requiring exact adjustment of the filter frequency to the ribbon fre- 
quency, constant control of the latter at a fixed value, and exact 
similarity of all ribbons so that they could be interchanged. A 
simpler method of using an undamped ribbon was to increase its 
tension so as to raise its natural frequency to 10,000 or 11,000 cycles 

/ 1 

FIG. 4. System for producing two images of the one 
ribbon: the prisms PI and PZ have been added to the 
system of Fig. 2 between the second objective and the 
cylindrical lens, and two slits are used at M 2 instead of 

per second. While for the most part satisfactory, it was found that, 
on sudden transitions between silence and loud sounds, faint traces 
of resonance would color the reproduced sounds in an indistinct but 
characteristic way. 

Both in the laboratory and in practice damping was attempted by 
placing a drop of "tavote" or oil at the points of attachment of the 
ribbon to its mounting. Such damping required extreme care, and 
the results were far from uniform. Important changes in the fre- 
quency characteristic resulted from small changes in the temperature 
of the ribbon and in the amount of oil used. The use of this method of 
damping in the Leningrad studios of Soyuzkino, however, has shown 
that the failure to eliminate the resonance peak completely merely 


colors certain speech consonants in a way that does not spoil the 
quality appreciably. 

It is also important that the deflection of the ribbon be propor- 
tional to the strength of the actuating current for all current strengths. 
Calculation shows that a change in this proportionality is possible 
only at current densities great enough to lengthen the ribbon by 
heating it and stretching it. Such changes in length are avoided 
by fixing the ribbon to its mounting through a spring. The mea- 
sured amplitude characteristic that results is entirely satisfactory. 

Any way by which a given value of current can be made to produce 
a larger effect on the film obviously will improve the modulator, 
particularly by permitting the use of smaller currents and thus avoid- 
ing overheating, straining, or breaking the ribbon. Such an increase 

FIG. 5. (Upper) Path of the light 
beam through the trapezoidal prism 
Pi of Fig. 4. 

FIG. 6. (Lower) The two sound 
tracks recorded by the system of 
Fig. 4. 

can be obtained either by adding a second ribbon, actually or opti- 
cally, or by achieving further enlargement in the optical system. 
The addition of a second material ribbon was tried and rejected be- 
cause of the precision required in mounting the two ribbons to insure 
that they were exactly spaced and that their images were precisely 
focused on the film. Furthermore, overmodulation caused clashing 
and short-circuiting. 

It was therefore decided to obtain optically two images of the one 
ribbon. A duplicator for that purpose was devised by adding two 
prisms to the system shown in Fig. 2, between the sectmd objective 
and the cylindrical lens, using two slits instead of one in M 2 , as shown 
in Fig. 4. Rays passing through one of the slits pass through the 
trapezoidal prism PI, as shown in Fig. 5, and produce an inverted 



[J. S. M. P. E. 

image on the film. Rays from the other slit, passing through the 
rectangular prism P 2 , produce an uninverted image, this prism serving 
merely as a compensator so that both bundles of rays receive the 
same optical treatment except for the inversion of the one. 

In use, such a system produces two sound tracks on the film, as 
shown in Fig. 6. Since the system brings the two images to a focus 
one above the other on the film, one sound track appears to lag with 
respect to the other by an amount equal to the width of the slit image, 
0.02 mm. By changing the forms of the prisms the two images, 
direct and inverted, can be placed side by side at the same height, 
but such prisms are inconvenient and complicated to construct. 
The same result can be accomplished more simply by constructing 
the cylindrical lens so that to either the upper or the lower half of 

100 1000 10000 

FIG. 7. Variation of amplitude of reproduc- 
tion of a pure frequency recorded by two images 
of the same ribbon, and reproduced by a device 
using a single ribbon. 

its flat side is effectively added a thin prism through which the rays 
from one of the slits pass and are thus deflected to the same height 
on the film as the rays from the other slit. 

It can be shown that if a pure frequency is recorded by two images 
of the same ribbon, one lagging with respect to the other, and the 
record is reproduced by a device using a single slit, the resulting 
sound will have the same frequency as the original, but will differ 
from the latter in amplitude and phase. The extent of the differences 
will depend upon the frequency of the sound and the relative devia- 
tion of the two tracks. In the case of a recorder in which the lag is 
0.02 mm. and the velocity of the film is 450 mm. per second, the 
reproduced amplitude falls off with the frequency, as shown in Fig. 7. 

To eliminate this relative deviation and its effects entirely, the 

Mar., 1934] 



optical duplicator diagrammed in Fig. 8 was designed. The dupli- 
cator proper, L 2 , is composed of two prisms, / and //, whose surfaces 
are in optical contact over the section back of the line AB, but are 
separated by an air space in front of the line. Thus, rays passing 
the ribbon N and proceeding through the objective LI and the slit 
P are divided in two parts, 1 and 2, at the dividing line AB. Part 1 
passes through the cemented section and is reflected downward 
by prism 77 to form the image NI". Part 2 is reflected upward by 
the surface of /, which is separated by air from II, and subsequent 

4\ , .8 

FIG. 8. (Upper) Optical duplicator for eliminat- 
ing the lag illustrated in Fig. 6. 

FIG. 9. (Lower) The two types of record pro- 
duced with the optical duplicator of Fig. 8. 

reflection from the two top surfaces of / sends the rays downward to 
form the image N 2 ", the inverse of NI". It is evident from the con- 
struction of the prism system that the two images will be precisely 
aligned. Furthermore, the quality of the images will be impaired no 
more than they would be if the rays had passed through a flat glass 
plate having parallel surfaces perpendicular to the direction of the 
rays, since all the rays pass through the same amount of glass, and 
enter and leave the glass at the same angles in the case of the prisms 
as they would in the case of the plate. 



[J. S. M. P. E. 

According to which end of the slit the ribbon is opposite, its 
two images on the film will either always overlap or always remain 
separate, producing the two types of record shown in Fig. 9. When 
the direct image A and the reversed image B overlap, they produce a 
record somewhat like what would be produced by a ribbon that be- 
came thicker and thinner in accordance with the sound current 
actuating it. 

The efficiency of the recorder was still further increased by using 
new objectives designed for greater magnification. Special im- 
mersion objectives designed by us for use with a ribbon damped by 
petroleum oil were manufactured by VOOMP. * The general scheme 

FIG. 10. Special immersion objective designed for use 
with a petroleum-damped ribbon. 

of such an objective, 18 mm. long and 11 mm. in diameter, is shown 
in Fig. 10. Not only is its magnification double that of the old ob- 
jective (20x instead of lOx) but it is improved in regard to spherical 
and chromatic aberration and satisfactorily meets the sine condi- 
tion. Taking all matters into account, the objective appears to 
produce images of good quality in a field of 12 mm., and hence within 
an object field of 0.6 mm. 

The next improvement was a combination of a 20x objective, im- 
mersional or non-immersional, with the duplicator first described. 
In this combination the slit M z was moved from in front of the 
duplicating prisms to a position behind them, and thus its construc- 
tion could be simplified to provide only one slit instead of two. The 

* All-Russian United Optico-Mechanical Industries. 


optical systems for use with oil and with air objectives differ only 
in the dimensions of their parts. 

With the additional magnification thus attained it is possible to 
reduce the amplitude of vibration of the ribbon, and hence in turn to 
reduce the current actuating it to about a quarter of its former value. 
Furthermore, the reduction of the required amplitude permits the 
use of a ribbon only one-fourth as wide, and thus its tension can be 
reduced to effect a still further reduction of current. 

The use of highly magnifying objectives complicated considerably 
the problem of designing the pole pieces of the magnet in which the 
objectives must be mounted. In the previous lOx optical system the 
distance between the outer lens and the ribbon was about 11 mm., 
readily permitting a satisfactory construction for the pole pieces. 
With a 20x air objective, the distance between the outer lens and the 
ribbon must be only 0.7 mm., and to accommodate the objective in 
this position the pole pieces must be given a peculiar form such that 
the whole flux passes through a very narrow strip of material. The 
immersional objective already described, therefore, was so designed 
as to make the distance between the ribbon and the first lens as great 
as practicable. The separation of 1.5 mm. actually achieved makes 
possible a far better magnetic circuit. 

Finally, the use of an achromatic lens to double the transmitted 
light, and the modification of the objective so as to double the dis- 
tance between the cylindrical lens and the ribbon, increased the en- 
largement even more, while still retaining a sufficient intensity of 
light in the image to permit satisfactory recording on positive film. 
The combined use of the 20x objective, the additional doubling lens, 
and the doubled distance provide a total magnification of 80-fold. 
With such magnification the current actuating the ribbon needs not be 
larger than 20 or 30 milliamperes, allowing greater freedom in the 
placement of microphones and permitting the use of recording ampli- 
fiers of considerably lower gain and lower power output potentialities. 

It is generally appreciated that in order to attain the best artistic 
effect in reproducing sound, background noise must be eliminated, 
and that such noise is produced by the undarkened portion of vari- 
able-width sound tracks and by portions of low density on variable- 
density sound tracks. The noise can be reduced by darkening those 
sections in proportion to the envelope of the recorded vibrations. 
To that end two methods, one wholly electrical and another partly 
mechanical, were developed. 



[J. S. M. P. E. 

In the former the sound currents are divided (Fig. 11), one portion 
passing through a stopping condenser in series with the ribbon, and 
the other through a copper-oxide rectifier and simple low-pass filter 
to the ribbon. The condenser in the first circuit has a capacitance 
of about 1000 micromicrofarads, and the circuit freely passes currents 
of frequencies higher than 40 or 50 cycles per second. Hence the 


FIG. 11. (Upper) Divided electrical circuit for noise 

FIG. 12. (Center) Negative record produced with the 
electrical noise suppressing system of Fig. 11. 

FIG. 13. (Lower) Noiseless recording produced with 
the optical duplicator. 

ribbon vibrates not only in accordance with the sound currents sup- 
plied by the first circuit, but at the same time shifts its position in 
accordance with the envelope of those currents supplied by the second 
circuit. When no current is passing, the image of the ribbon is near 
the edge of the sound track, almost completely covering it, and when 
the current is small most of the sound track is darkened, producing a 


negative print such as that shown in Fig. 12. When a larger current 
passes, the image of the ribbon shifts farther from the edge of the 
track and exposes more of the film. 

By such an arrangement, the ribbon is obliged to conduct twice the 
current that it would otherwise be required to carry, a disadvantage 
when the current is large. To avoid it, the ribbon is mounted so 
that one edge of its image normally falls at the center of the sound 
track, and biasing current is supplied by a battery through a vari- 
able resistance such that the current is sufficient to deflect the ribbon 
so that its image falls at the edge of the sound track as before. By 
then permitting the current from the rectifier to oppose the biasing 
current, neutralizing it when the sound current reaches a maximum, 
the ribbon is required to carry no more current than in normal record- 
ing during loud passages. 

The fact that the sound track is modulated only at the extreme edge 
by currents of small amplitude is a disadvantage, for in poorly ser- 
viced theaters, where the optical system of the reproducer is im- 
properly placed laterally with respect to the film, or incorrectly 
focused, distortion of the weak sounds may result. This disadvan- 
tage is obviated by the use of the duplicator, which requires no change 
in the noiseless recording system, and which centers the modulated 
portion of the sound track regardless of the current strength, as 
shown in Fig. 13. 

In the other method of noiseless recording the voice currents are 
divided between the ribbon and a small galvanometer (Fig. 14). 
A small vane is attached to the needle of the galvanometer, the motion 
of the needle being damped by a wire attached to it immersed in 
glycerine. The vane is thus moved from side to side in accordance 
with the average strength of the sound current, after a delay due to 
the damping. By focusing an image of the vane on the film at such a 
distance from the image of the ribbon as to allow for the delay of re- 
sponse of the vane, and adjusting the amplitude of its motion by 
means of a resistance in series with the galvanometer, the device can 
be made to produce a positive in which the envelope of the modulated 
sound track is darkened. The galvanometer requires at the most 
one-tenth of the total current for its operation. One embodiment of 
this device is shown in Fig. 15. 

We have been no exception to the general rule that every one who 
works on the problems of the sound film must suffer and recover from 
the children's disease of experimenting with Kerr's cell. We did not 



[J. S. M. P. E. 

experiment long with the system, nor intend it for general use, be- 
cause of its well-known basic faults : small light intensity, large har- 
monic output when inaccurately controlled, the necessity of high 
voltages, expensive optical systems and the use of nitrobenzol, and 
the danger of freezing at low temperatures. The system has been ex- 
tensively used in Germany, where those shortcomings have been over- 
come by indirect methods not available to us, such as the use of large 
prisms, and film of very fine grain and high sensitivity. 

FIG. 14. (Upper) Electromechanical system of noise 

FIG. 15. (Lower) One embodiment of the electro- 
mechanical system of noise suppression of Fig. 14. 

Work with a flashing lamp was also undertaken, in the hope of 
achieving a simpler and more compact form of recording apparatus 
than was possible with other means. It was found possible to use 
such a lamp for recording on negative film, but the lamp could not be 
made bright enough to record the necessary densities on positive film. 
It was difficult to produce a sufficiently large glowing surface that 
would radiate over a wide enough angle to permit proper employment 
of the necessary optical systems. 

Mar., 1934] 



We investigated the dependence of the brightness of the lamp on 
the voltage supplied to it by exposing negative film to the lamp while 
the voltage was varied. Previous measurements of the relation 
between the density of the film and its exposure then enabled us to 
determine the variation of the lamp brightness with the voltage. The 
intensity of the light radiated from the central portion of the light 
source was also measured photographically (Fig. 16). The source 
was imaged by an objective on the entrance opening of a second 
objective, a diaphragm of which occluded all but the central portion 
of the image. The second objective imaged the exit opening of the 

FIG. 16. (Upper) Measurement of intensity of radiation from central 
portion of flashing lamp. 

FIG. 17. (Lower) Intensity of radiation of central portion of flashing lamp 
at various angles. 

FIG. 18. (Right) The flashing lamp used in the experiments. 

first objective on a negative film. Thus measurements of the dis- 
tribution of density of the developed negative could be used to deter- 
mine the radiation of the central portion of the source at various 
angles (Fig. 17). The flashing lamp constructed in this laboratory 
was of the type shown in Fig. 18, and was employed in conjunction 
with the optical system diagrammed in Fig. 19. 

These experiments with various types of modulators, and ex- 
amination of the available literature on the subject, have convinced 
the author that the best sound recording systems are those in which 
mechanical vibrations are used to modulate the light beams, since 
such systems provide the most intense illumination and are generally 



[J. S. M. P. E. 

the most compact. The author hopes he may be pardoned for dar- 
ing, with some excusable partiality, to compare from his standpoint 
the merits of the various light modulator systems that have recom- 
mended themselves. In this comparison it is assumed (a) that the 
recording is on regular positive motion picture film, (b) that all 
frequencies from 50 to 10,000 cycles per second are recorded, and (c) 
the image of the slit on the film measures 0.02 mm. by 3 mm. 

Table I roughly rates the extent to which the most important re- 
quirements are met by five different modulators: Western Electric 
(W), RCA (R), the author's (Ch), Kerr's cell (K), and the flashing 
lamp (L) . Where a requirement is completely fulfilled, the sign + is 
entered; where not, the sign ; and where a modulator fulfills the 
requirement poorly in comparison with other modulators, the sign =t . 


FIG. 19. Optical system employed with the flashing 
lamp of Fig. 18. 

(1) The luminous intensities of modulators W, R, and Ch are 
sufficient for positive film; Kerr's cell only barely supplies enough 
light when supplemented with large Nicol prisms, special preparations 
of nitrobenzol, etc., as by Klangfilm Tobis. Flashing lamps can not 
be made to supply enough light and at the same time fulfill the other 
requirements for positive film. 

(2) All the modulators possess the requisite frequency character- 
istics except the flashing lamp, the characteristic of which is in- 
consistent and indicates the presence of hysteresis. 

(3) In the first three modulators the amplitude of response is pro- 
portional to the amplitude of the actuating signal, within the re- 
quired range. In the last two that is true only under very special 
conditions of battery voltage and the like. 

(4) All the modulators require that the current supplied to them 
be amplified. Differences in the degree of amplification required are 



Characteristics of Five Different Types of Modulator 

W R Ch K L 

( 1) Light Intensity + -f + =*= 

( 2) Frequency 

Characteristic + -f- + -f == 

( 3) Amplitude 

Characteristic + + + =*= =*= 

( 4) Amplification 

Required + + + + + 

( 5) Working 

Voltages + + + 

( 6) Type of Record 

Produced + + 

( 7) Distortion on 

Overmodulation + -f- == 

( 8) Influence of 

Temperature + + + + 

( 9) Complexity + + + + 

(10) Noiseless 

Recording + -f- 

(11) Extent of 

Modulation + + 

of no great importance in practice, requiring merely different numbers 
or kinds of vacuum tubes and different amplifier circuits. Apparently 
the Western Electric System requires the least energy for actuating 
the ribbon. 

(5) The first three modulators use very low voltages throughout. 
Kerr's cell and the flashing lamp require the use of inconveniently 
high voltages close to the modulator. 

(6) Only the RCA and the author's modulators produce variable- 
width records, the photographic manipulation of which is much 
simpler than that of variable-density records. As already described, 
the author's system can be used for either method of recording without 
change of construction. 

(7) In spite of the argument that records can be made without 
overmodulating them, the danger of doing so can not be disregarded. 
In the Western Electric System overmodulation produces a short- 
circuit. In the RCA and the author's systems overmodulation causes 
the movements of the ribbon image to exceed the limits of the sound 
track, and hence the reproduced sounds are distorted. The distor- 
tion takes the form of added higher harmonics, however, which do 

172 A. F. CHORINE 

not entirely spoil the sound. In Kerr's cell overmodulation changes 
the direction of the current, greatly distorting the record, and can 
be detected and avoided only by special devices. In the flashing 
lamp, distortion due to overmodulation varies with the characteristic 
of the lamp. 

(8) All the modulators work well at all temperatures except those 
employing Kerr's cell, which will not operate at low temperatures 
without special arrangements for controlling the temperature with a 

(9) As so far developed, most modulators are not unduly com- 
plicated, except again Kerr's cell in any form that provides enough 
light for recording on positive film. For simplicity the Western 
Electric modulator has not been surpassed. 

(10) The RCA and the author's systems seem to be those in which 
noiseless recording methods can be most satisfactorily applied, but 
since the last word has not yet been said in the matter, the ratings 
given here must be regarded as open to question. Noiseless record- 
ing with the variable-density systems always distorts the reproduction 
because the amount of light transmitted by any portion of the posi- 
tive sound track, when the sounds are reproduced, is not proportional 
to the amount of light required to record that portion of the sound 
track, over the range of density that is encompassed. 

(11) It can be said, in general, that the extent of modulation is 
greater with variable-width than with variable-density systems. 

A glance at the table now shows that, from the standpoints adopted, 
the systems employing Kerr's cell and the flashing lamp are the least 
satisfactory, and the RCA and the author's systems are the most. 
The comparison, of course, affords only a rough idea of the situation 
today, and different points of view might lead to different conclu- 

It is interesting to notice that this work, undertaken primarily 
to develop apparatus for sound picture use, has borne fruit in an in- 
strument useful not only in that field but as a tool for scientific in- 
vestigation. The light modulator constitutes an ideal oscillograph. 
Only modulators producing variable-width records can properly be 
considered as microscopic oscillographs, as variable-density records 
can be used as oscillograms only with difficulty. 

(To be concluded in the April issue of the Journal) 


Prepared under the Supervision of the 


Summary. The Visual and Sound Test Reels, developed under the supervision of 
the Projection Practice Committee, were described fully in the August, 1933, issue of 
the Journal. The reels are now available, and consist of two sections, each approxi- 
mately 500 feet long. They are designed to be used as precision instruments in thea- 
ters, review rooms, exchanges, laboratories, and the like, for testing the performance 
of projectors. Either or both sections may be obtained on order accompanied by a 
remittance of $37.50 for each section, from the General Office of the Society, at 33 West 
42nd Street, New York, N. Y. 


The test reel consists of two separate sections of standard 35-mm. 
film, each section being approximately 500 feet long : (1) picture, and 
(2) sound. The two portions can, and preferably should, be used 


To enable the projectionist to check his equipment for optical and 
sound defects within the limited time usually available prior to the 
opening of performance; and to aid in eliminating such defects, 
thereby helping him to maintain a high standard of projection. 


The picture section consists of various test targets, * each preceded 
by a title stating the purpose for which it is intended, arranged in the 
following order : 

(a) Small diamonds and vertical bars arranged alternately in rows for checking 

(6) Small squares arranged diagonally across the frame, for checking picture 
jump and picture "weave." 

* Illustrated in the August, 1933, JOURNAL, p. 90-94. 



(c) Fine vertical lines closely spaced, for checking marginal and radial aber- 
ration (imperfection) of objective (projection) lens. 

(d) Fine horizontal lines closely spaced, for checking marginal and radial aber- 
ration (imperfection) of objective (projection) lens. 

(e) Small squares for checking best focal position of objective lens. 

(a) To Check for Travel-Ghost. Set the shutter adjusting knob 
to its central position. Set the shutter approximately to the correct 
position. This permits advancing or retarding the shutter while the 
projector is running. White streaks appearing on the screen above 
or below the white objects indicate the presence of travel-ghost, 
which should be removed by manipulating the adjusting knob: 
streaks above indicate that the shutter should be advanced; streaks 
below indicate that the shutter should be retarded. Intermittent 
streaks indicate insufficient tension of shutter-cam spring, excessively 
worn gears, or worn bearings. If the projection distance is great, an 
observer should be stationed near the screen, so as to detect faint 
travel-ghost that might be invisible from the projection room, and to 
signal to the projectionist whether the streaks are "up" or "down" 
and when they are entirely eliminated. 

(b) To Check for Picture Jump and Picture Weave. The picture 
jump is measured by placing a ruler against the screen and measuring 
the amount of movement of one of the squares in a vertical direction. 

If the projector is in first-class condition, and the intermittent 
movement and the picture gate are properly adjusted, the picture 
jump should not exceed the values given in the following table (one- 
third of one per cent of the picture height) : 

Vertical Movement Height of Picture 

(inch) (feet) 

Va 3 

V< 6 

3 A 9 

Va 12 

5 A 15 

3 A 18 

Vs 21 

An excessive amount of picture jump may be due to one or more of 
several causes, namely: 

(1) Insufficient or excessive gate tension. 

(2) Worn intermittent sprocket teeth. 

(3) Intermittent sprocket of incorrect dimensions. 


(4) Dirt accumulated upon face of sprocket. 

(5) Improper adjustment of intermittent movement. 

(6) Vibrations transmitted by projector motor. 

(7) Bent shutter shaft or unbalanced shutter. 

(8) Insecure foundation for projectors. 

(9) Loose lens holder or lens elements. 

(10) Improper fitting of intermittent sprocket on shaft. 

(11) Worn or damaged test-reel sprocket holes. 

The picture weave is measured by placing a ruler against the screen 
and measuring the amount of movement of one of the squares in a 
horizontal direction. If the projector is in first class condition, the 
weave should not exceed one-third of one per cent of the picture width. 
The table provided above for picture jump may be used, substituting 
"horizontal" for "vertical" in the first column, and "width" for 
"height" in the second column. Excessive picture weave may be 
caused by: 

(1) Insufficient tension on lateral guide rollers. 

(2) Misadjustment, sticking or cut lateral guide rollers. 

(3) Improper adjustment of intermittent sprocket. 

(4) Improper fitting of intermittent sprocket on shaft. 

(5) Worn or damaged test-reel sprocket holes. 

(c, d) To Check for Lens Aberration. All commercial projection 
lenses have defects called aberrations, which are corrected to a greater 
or lesser degree depending upon the type of lens and the quality of 
the workmanship and materials used in its manufacture. Blurring 
of any portion of the targets indicates that the lens has not been fully 
corrected. No adjustment can be made by the projectionist to 
correct such a condition; but the targets should be used to make com- 
parisons between lenses, especially before purchasing new ones. 

(e) To Check Best Focal Position of Lens. The focal adjustment 
of the lens is best when the greatest possible number of squares is 
sharply and clearly defined on the screen. The rows of squares are 
numbered vertically and horizontally for the purpose of locating de- 
fects and for comparing different optical systems. 


The sound section of the test reel consists of an assortment of sound 
recordings to be used for detecting faults in the reproduced sound and 
for checking the performance of the projectors and the associated 
sound equipment. Two sound tracks, each approximately 500 feet 


long, are printed near the left and the right margins of the film, with 
the starting points at the opposite ends of the film. Such an arrange- 
ment permits the two sound tracks to be used without rewinding the 
film. In order to use the second sound track, the reel may be removed 
from the lower magazine, turned so as to present the emulsion side to 
the light source, placed in the upper magazine, and threaded in the 
usual manner. The sound recordings are as follows : 

(a) Buzz track (a track that is practically the reduced image of the rungs of 
a ladder) for checking the position of the scanning beam relative to the sound 

(b) 6000- Cycle and 9000-cycle note for checking the focus and rotational 
adjustment of the sound optical system. 

(c) Selected frequencies, including 50, 100, 200, 300, 500, 1000, 2000, 3000, 4000, 
5000, 6000, 7000, 8000, 9000, and 10,000 cycles for ascertaining the over-all out- 
put characteristics of sound heads and amplifiers. This track was recorded at a 
constant level in order to avoid voltage calibration when a volume indicator is 
used (i. e., the volume indicator should always indicate the same reading). In 
listening, the 1000-cycle note will sound louder than the others, because the nor- 
mal ear is more responsive to notes of that frequency than to higher or lower 
notes. This track may be used to check the range of frequency covered by the 

(d) Vocal recording for checking intelligibility of speech and theater rever- 

(e) Piano recording of sustained notes for checking "flutter" and "wows." 
(/) Orchestral recording for checking naturalness of reproduction. 

(Tracks a, b, and c are on one margin of the film. Tracks d, e, and/ are on 
the opposite margin.) 

(a) Buzz Track. For checking correct alignment of film with the 
sound gate. 

(i) Be prepared with proper tool to make adjustment of lateral 
guide rollers of sound gate, if necessary. 

(ii) Project film with amplifiers in operation and fader up. 

(iii) When the 1100-cycle (higher) note is heard, the film is pass- 
ing the scanning beam too closely to the sprocket hole margin. 

(iv) When the 300-cycle (lower) note is heard, the film is passing 
the scanning beam too closely to the picture margin. 

(v) When both notes have been eliminated by properly adjusting 
the lateral guide roller (or by adjusting the optical system, where such 
adjustment is intended) , correct film travel may be assumed. Adjust- 
ment of the optical system should not be undertaken unless the sys- 
tem has been designed to be adjusted and suitable tools are available. 
When one or both notes are heard intermittently, the film weave is 


excessive, or the optical system improperly focused or not standard. 
Incorrect relative alignment of the projector head and the sound head 
may also cause film weave. 

(b) 6000-Cycle and 9000-Cycle Constant- Frequency Tracks. These 
are for the purpose of focusing and checking the rotational position 
of the scanning beam. The 9000-cycle track consists of parallel 
lines 1 mil wide and spaced 1 mil apart, facilitating the very closest 
adjustment. The 6000-cycle note is provided for use when a volume 
indicator is not available and when the 9000-cycle note is inaudible 
because of the limited frequency range either of the reproducing equip- 
ment or of the ear of the listener. 

When maximum volume is attained at both frequencies by carefully 
adjusting the focus, it may be assumed that the optical system is 
correctly positioned for best results ; that is, that the scanning beam is 
sharply focused and is at right angles to the direction of film travel. 
Adjustment of the optical system should not be undertaken unless 
the system has been designed to be adjusted and suitable tools are 
available. The type of optical system having the light slit fixed with 
reference to the machined base of the optical unit will not permit 
rotational adjustments to be made. 

(c) Selected Frequencies. This is used for checking the over-all 
output characteristics of the sound head and the amplifier (using 
a volume indicator), and for listening tests to determine the range of 
frequency covered by the reproducing equipment. This track is 
also useful for locating the causes of rattling or buzzing of loud speak- 
ers. Among the various frequencies there may be one that will cor- 
respond to the natural frequency of the loose wire or part, causing it 
to vibrate and thus enabling it to be located. 

(d) Vocal Recording. This provides a good test for frequencies in 
the middle of the range, which exert the greatest influence on the 
intelligibility and clarity of speech. 

(e) Piano Recording. This is intended for checking and deter- 
mining orally the extent of "wows" or "flutter," which are more easily 
detected in the sustained notes. Any factors that cause a variation 
in the speed at which the film passes the scanning beam, such as worn 
sound sprocket teeth, off-center sound sprocket, accumulation of dirt 
or wax on the sound sprocket, improper sound gate tension pad, or 
uneven tension on take-up, will cause "wows" or "flutter." A loose 
exciting lamp or machine vibration may also cause "flutter." 

(/) Orchestral Recording. This is to provide a test of the natural- 


ness of reproduction, which will depend on the range of frequency 
covered by the reproducing equipment. This recording contains 
notes ranging from those of the lowest tuba and double-bass up to 
the very high overtones of the brass instruments. The metallic 
quality of certain instruments, such as the wire brushes, should be 
particularly noticeable. 


The test reels are instruments of precision, and should be handled 
with the special care befitting such devices. Accordingly, before 
threading the films into the projector, the entire track through which 
the film is to pass should be carefully cleaned and examined to avoid 
any likelihood of film damage. It is recommended that between 
showings of the test reel, it be wound tightly on a metal reel having 
a large hub, and stored in a metal reel-can which should be made as 
air-tight as possible (for example, by using adhesive tape). 


At a meeting of the New York Section of the Society, held at New York on December 
13th, the following address was delivered by Mr. Barton through the medium of a disk 
record of Victrolac 1 recorded through a high-fidelity recording system and reproduced 
through a properly compensated high-fidelity reproducing system. Following the re- 
production of the record, the meeting was opened to a general discussion of the system. 

Mr. Chairman and Gentlemen: 

It is our privilege to demonstrate some experimental records which 
have been made through a high-fidelity recording system, reproduc- 
ing them through a properly compensated high-fidelity reproducing 
system. The records are of the conventional lateral-cut variety; 
that is, the vibrations representing the sound waves pass from side 
to side in a horizontal plane. 

I do not intend to present a technical paper on the subject 
of high-fidelity lateral recordings, or burden you with the pres- 
entation of curves analyzing the various parts of the system. Suf- 
fice it to say that the over-all system, from the original sound enter- 
ing the microphone to the reproduced sound leaving the loud speaker, 
is substantially flat. Certain units in the system have characteris- 
tics that are anything but flat. However, for each variation that 
occurs in one part of the system there is a compensating variation 
in the opposite sense in some other part of the system; and these when 
algebraically added will produce a substantially flat over-all system. 
An example may be found in the recorder, which at 100 cycles may 
be down approximately 12 db. To off -set this the reproducing amp- 
lifier will be approximately 12 db. up at 100 cycles. 

The records to be played are short extracts made during regular 
recording dates, which will account for the fact that in most cases 
they are incomplete, at least in so far as the beginnings and the ends 
of the selections are concerned. 

* Demonstrated before the New York Section, December 13, 1933. 
** RCA Victor Company, Camden, N. J. 


180 F C. BARTON [j. s. M. P. E. 

These records, while not presented as indicating that perfection 
has been achieved, do indicate that a degree of fidelity far in ad- 
vance of that normally attained is possible with the lateral recording 
system. Advances in technic of considerable magnitude have been 
achieved even since these records were made, and it is hoped that 
sometime in the not too distant future another demonstration of a 
more finished nature may be made. 


1 BARTON, F. C.: "Victrolac Motion Picture Records," /. Soc. Mot. Pict. Eng. 
XVIII (April, 1932), No. 4, p. 452. 


MR. EVANS: What is the material of the record? 

MR. BARTON: Victrolac. It is a new synthetic resin, which has been found 
adaptable to the manufacture of high-grade records. The motion picture records 
were dyed black. 

MR. EVANS : What is the range of volume between ground noise and maximum ? 

MR. BARTON: Some improvements have recently been made that have de- 
creased the background noise considerably. I must revise my figures before 
making definite statements. 

MR. HUNT: The records were started without any noise whatsoever; but it 
seemed that after the start a certain amount of surface noise could be heard. 

MR. BARTON: Because of the abrupt start of the music, in most cases the 
pick-up was put on the record and the sound faded in. The background noise of 
some of the records was rather considerable. Improvements have been made, as 
I said when speaking from the record. The surface noise has been dropped 12 db., 
so that it is now quite in the background. 

MR. MARMON: Is the reduction in surface noise due entirely to the material of 
which the record was made, or other contributing factors? 

MR. BARTON: There are many factors, principal among them being the 
processing of the wax. 

MR. EVANS : What are the speed, pitch (the number of grooves to the inch) , the 
depth of the cut, the width of the cut? What is the commercial status of the 
records? Is the development directed toward home entertainment or for use in 
connection with pictures as well? 

MR. BARTON: The pitch is approximately 100 grooves per inch, which is the 
normal commercial pitch. The speed is 78 rpm., the records having been cut on a 
standard recording machine. The groove is narrower than the standard groove 
by a couple of thousandths of an inch. The development is directed primarily 
at this time toward home use; it will probably branch out. It is felt that the 
record for home consumption has stood still for quite a long time, and that it is 
time something were done about it. 

MR. EVANS: Is a special reproducer required for playing in the home, or can 
one play the records with the present reproducing equipment? 

MR. BARTON: A special reproducer is required. 

Mar., 1934] DlSK RECORDS 181 

MR. EVANS: Have you an instrument for playing back from the wax? 


MR. YAGER : How does the volume range of the records compare with that of 
the ordinary record? 

MR. BARTON: Volume range is entirely a matter of surface noise to signal 
ratio. In the records of the Philadelphia Orchestra, the surface noise was down 
far enough to take advantage of the increased range. The records were not moni- 
tored. If you know the volume range of the Philadelphia Orchestra, then that is 
the range of the records. 

MR. TASKER: By monitoring I suppose you mean the adjustment of gain 
during recording? 


MR. YAGER: Is the deflection of the record truly lateral or at an angle? 

MR. BARTON: It is truly lateral. The chief difference between these records 
and the commercial records as they are sold now is that the recording systems are 
flat to 9500 cycles and the groove is of different size. The recorders are up to 
90 per cent of normal at 10 kc. The low-frequency attenuation of the recorders is 
somewhat different from that of ordinary commercial recorders, but that merely 
indicates that the reproducer amplifier characteristics should be the inverse of the 
recorder low-frequency attenuation. 

MR. EVANS: Are the high frequencies compensated for as well as the low? 

MR. BARTON: Yes. The high-frequency response is up about 4 db. in record- 
ing, and is correspondingly attenuated in reproduction. 

MR. POPOVICI: What is the maximum level that can be impressed on the re- 
corder without cutting over? 

MR. BARTON: Do you mean in thousandths of an inch? 

MR. POPOVICI; No, in decibels. 

MR. BARTON: I would rather avoid the question of levels. If you would 
like the answer in terms of physical measurements, we can swing the recorder 
stylus a maximum of about 4 mils; that is, 2 mils on either side of the center. 

MR. TASKER: I suppose you refer to the amplitude at relatively low frequen- 

MR. BARTON : Relatively low frequencies. The recording system has practi- 
cally constant amplitude characteristics from about 800 cycles down, and constant 
velocity characteristics from 800 cycles up. 

MR. Ross: Does the tone-arm follow the groove freely or is it screw driven? 

MR. BARTON: It is free; but no trouble is experienced from jumping the 

MR. TASKER: We can all agree to the last statement, having heard the tre- 
mendous range of the records without any jumping of grooves. 

MR. MANHEIMER: Is there a counterbalancing arrangement on the reproducer 

MR. BARTON: Yes, although the pick-up is very light in itself. 

MR. DIAMOND: Is an ordinary needle used for the reproducer? 

MR. BARTON: We use a diamond. With the very light pick-up the weight of 
the set-screw for clamping the needle would alone be entirely out of the question; 
in fact, the whole moving system in the pick-up probably weighs less than a steel 
needle considerably. 

182 F. C. BARTON 

MR. Ross: What other equipment contributes to the quality, aside from the 
recording process? 

MR. BARTON: The standard high-fidelity RCA Photophone film reproducing 
system is used. At the input of the system is a small amplifier, which has the 
necessary compensation to differentiate between films and disk; in other words, 
it has the necessary characteristics to bring the disk recording out flat when 
coupled to this Photophone reproducer. 

MR. MANHEIMER: Is there only one speaker back of the screen? 

MR. BARTON: Yes, a single speaker. 

MR. DROSATTI: How many times can you play the records? 

MR. BARTON: I don't know; a great many times. Some day when I have the 
time I shall cut some that stay in the same groove, and play them to death. It 
would take a long time to play a complete record enough times. 

MR. Ross: I should like to hear the recorded speech again, and afterward 
hear you repeat part of the speech in the same tone. Your voice seemed somewhat 
more formal through the microphone. (Mr. Barton did as Mr. Ross requested.) 

MR. EVANS : How far did you stand from the microphone when the record was 

MR. BARTON: About 2 l / 2 feet. 

MR. KELLOGG: The record sounded to me as though you were in a much more 
resonant room than this when you made the speech over the record. Also you 
have been standing about 10 or 12 feet nearer us than the loud speaker. This 
increases the contrast between your natural and recorded voice. If you had stood 
farther back in carrying out the test requested by Mr. Ross the naturalness would 
have been increased. 

MR. BARTON: I say there have been advances since the record was made. 
You see, I am supposed to be talking. The advance took place just prior to the 
time this particular speech record was made. 

MR. KELLOGG: Then the words "these records" in your recorded speech refer 
only to the musical records? 



At a meeting of the New York Section of the Society, held at New York on December 
13th, an open forum discussion of the subject of compensation was held. The dis- 
cussion follows: 

MR. TASKER : We have heard a number of examples this evening of sound that 
was produced through recording and reproducing systems that were engineered 
to cooperate in some manner or other. The meeting is now open for remarks on 
the manner of engineering the two systems with respect to each other. 

MR. EVANS : Quite a long time ago, two or three years at least, it was suggested 
that the recording system be made to compensate for the theater characteristics. 
I advocated rather strongly to the Sound Committee that that be done, but the 
majority of the Committee seemed to feel otherwise. My reason for advocating 
it at that time was that the reproducing equipment of those days was rather 
deficient in both high and low frequencies, and it seemed as though some of the 
deficiency might be compensated for in the recording system. The majority of 
the Committee felt, however, that nothing of that sort should be advocated be- 
cause it would not be long before the reproducing equipment would be improved, 
and then anything that had been so recorded as to sound well on existing equip- 
ment would not sound so well on the improved equipment. The discussion con- 
tinued for a couple of years, but the recommendation was finally made that the 
recording system should be so designed as to produce a flat characteristic on the 
print, and that the reproducing equipment provide a flat characteristic from 
the print through the loud speaker. I do not know whether the Standards 
Committee has acted in the matter. 

MR. BATSEL: It has not been presented to the Standards Committee. 

MR. EVANS: It has, however, been formally adopted by the Sound Com- 
mittee. At about the time it was adopted the new reproducing equipment was 
becoming available. RCA was bringing out the high-fidelity system and Western 
Electric its wide-range reproducing equipment. Strangely enough, at about the 
same time some of the studios on the West Coast acquired the idea of compen- 
sating for the reproducing equipment during the recording; and in the spring we 
began to notice that some of the pictures coming from the Coast had been so 
compensated, with the result that we are now faced with a rather difficult situa- 
tion. According to recent figures, about 1000 theaters are now equipped with 
high-fidelity or wide-range systems mostly class A houses, I believe and it is 
very confusing at the present time to have some of the pictures compensated 
during recording for the older equipment: when these are played on the newer 
equipment they do not sound as well as when reproduced on the older type of 
equipment. In July Dr. Goldsmith asked the Sound Committee to consider the 
subject of standardizing the frequency characteristic and to try to induce the 


184 OPEN FORUM [j. s. M. P. E. 

various studios to adhere to such a standard, so that the recordings made by all 
the studios would sound alike on the same equipment. This problem will be one 
of the most important before the new Committee this coming year, and I hope 
some progress will be made in solving it. 

MR. KELLOGG: Did the compensation take the form of accentuating the bass 
or accentuating the high frequencies? 

MR. EVANS: More particularly the high frequencies. 

MR. McNAiR: It seems to me that the recommendation of the Sound Com- 
mittee that the recording of itself be flat, and the theater equipment also of itself 
be flat, is a sound point of view. It might be well to ask, though, whether any one 
knows the reasons why certain producers were led to make films that were not in 
accordance with that principle. (No answer.) 

MR. TASKER: It is true that compensation that raises the high frequencies 
above the normal level during recording, and decompensates to the same extent in 
reproducing, provides inherently a lower noise level than can be attained with the 
so-called ideally flat characteristics? 

MR. EVANS: I believe that that statement is true only for a limited amount 
of upward compensation of high frequencies during recording, with appropriate 
downward compensation, or decompensation, during reproduction. If we plot 
the curve of acoustic energy produced by the ideally flat reproducer through un- 
modulated film we should expect a noise-energy curve that would continuously 
rise toward the higher frequencies. If this 1 curve were then multiplied by the 
relative sensitivity of the ear for each frequency a new curve would result that 
would be a figure of merit for this system. If, then, the recording system is com- 
pensated upward to correspond to this latter curve, and the reproducing system 
decompensated downward to an equal extent, we should attain the lowest noise 
level that is possible. 

MR. YAGER: By increasing the high-frequency response in order to allow for 
the loss in compensation, the noise level of the entire system increases with the 
gain of the recording amplifier; and nothing is gained unless the noise level is 
entirely within the range of compensation. 

MR. EVANS: We must bear in mind when considering noise level, that the film 
is probably at the present time the controlling element. I do not believe that the 
system noise is as great as the film noise after the film has been run through the 
projector a few times. The noise level increases considerably after a few trips 
through the projector. 

MR. BATSEL: Our experience indicates that the point Mr. Evans and Mr. 
Yager have made is essentially correct. If it is necessary to compensate or 
equalize for losses in reproducers, the recording system would probably have to be 
peaked so much for frequencies as high as 9000 cycles that the swing for full 
modulation would be taken up by those frequencies instead of frequencies in the 
range contributing most to the loudness of the speech or music. The gain re- 
quired for reproduction would be increased to such an extent as to result in per- 
haps more noise than is produced by a flat system. In reproducing, there is a 
certain amount of extremely high-frequency hiss that would be lost by attenuat- 
ing the high frequencies, but much can be accomplished in limiting noise by having 
a reproducing system free of peaks. The noise is much less, as pronounced peaks 
at some frequencies may increase the noise without having much effect on apparent 

Mar., 1934] OPEN FORUM 185 

loudness. If the system is smooth, some of the hiss due to the film grain is elimi- 
nated by attenuating the highs. By carrying compensation to extremes the 
noise will be increased. 

MR. RICKER: In the average theater, or even in class A theaters, to what 
extent can a loud speaker reproduce film grain noise? 

MR. TASKER: Do you mean with respect to the frequency? 

MR. RICKER: The point I wish to make is this: that the loud speakers are not 
able to reproduce frequencies such as would be represented by the film grains 
that I find and measure under the microscope. I believe that many of you en- 
gineers charge ground noise against film grain when it should be charged to some 
other factor. 

MR. TASKER: I believe it is a fact that noise from film does extend over the 
whole frequency range; but it is obvious that any 500-cycle components, for 
example, could scarcely arise from film grain or even grain clumps, but must be 
due to irregularities of the base or variations of the sensitivity of the emulsion, or 
something of that sort. 

MR. RICKER : A careful examination of film after it has been used a few times 
and dried out would show reason enough for ground noise. There will be found 
scratches and translucencies that come into the base as well as in the emulsion; 
enough to be responsible for the increased noise. If the base is examined by re- 
flected light under a microscope, or by transmitted light, it will be found that the 
most perfect base is anything but constant in its refractive effect. In other 
words, the piece of film base, when magnified, looks like a piece of corrugated glass 
instead of a piece of flat glass. Nitrate base is more perfect than acetate base; 
but either, when stripped of emulsion, will provide plenty of ground noise if run 
through a sound head. 

[Note : Measurements of the ground noise of film base are given in the article 
by O. Sandvik, V. C. Hall, and W. K. Grimwood, on p. 83 of the February, 
1934, issue of the JOURNAL Ed.] 


Summary. A device proposed for use in all theaters as a standard method of 
changing from one projector to the other is described. The device is controlled by a 
constant frequency photographically recorded on the edge of the film a given distance 
from the end of the picture portion of a reel of film. Its purpose is to eliminate 
individual cueing and improve the presentation of talking motion pictures by en- 
abling the projectionist to make change-overs more exactly without visible marks on 
the screen, and to avoid to a large degree the film mutilation and increased fire hazard 
that results from the practice of individual cueing. 

The primary purpose of this paper is to outline the principles 
and methods of operation of an automatic sound and picture change- 
over device, intended for use in theater projection rooms and 
all other places where 35-mm. film is projected, using two or more 
projectors in sequence. 

The purpose of the device is to improve the presentation of 
motion pictures in enabling the projectionist to make change-overs 
more exactly and to eliminate the visual cues and individual cueing 
that result in the mutilation of film. It is controlled by a record, 
similar in appearance to a portion of a constant frequency sound 
track, recorded on the prints photographically between the row of 
perforations and the edge of the film at a certain distance from the 
end of the picture portion of the reel. 

Cueing Methods. Until about three years ago the entire problem of 
cueing for change-overs was a matter for the projectionist to solve to 
the best of his ability. That was a very undesirable situation, since 
practically every projectionist had his own way of cueing. Some 
scratched lines in the emulsion, some scratched crosses; some punched 
holes of a particular shape so as to distinguish theirs from the other 
holes previously punched by other projectionists. Others preferred to 
use click splices, and in order to distinguish theirs from the others, 
they used a certain color or shape or placed them in certain positions 
on the frame. Others wrapped a strip of tinfoil around the edge of 
the film, and in many cases an adhesive was employed which 

* Presented at the Fall, 1933, Meeting at Chicago, 111. 
** Coronado, Calif. 

186 * 


caused the film to crack, requiring that portion of the film to be 
removed. Others trimmed the edge of the film close to the sprocket 
holes, thereby weakening the film which also resulted in the later 
removal of that portion of the film. Others who had a little respect 
for their profession and the producers, chose to use china-mark- 
ing pencils of their preferred colors, so that the cue might be 
erased and not permanently mutilate the film. 

In the interest of improving change-overs and conserving the 
prints, the standard release print, which was made effective Nov. 1, 
1930, provided standard cue dots photographically recorded in the 
picture portion of the film for eliminating the individual cueing and 
improving the presentation of the picture. It is now definitely 
proved that the standard dots were not the remedy for the situation. 
The majority of the projectionists have returned to their old methods 
of cueing, apparently paying no attention whatever to the standard 
dots except when they are so conspicuous that they could not pos- 
sibly be missed, and even then sometimes they apply their own cues. 

An increasing number of projectionists are at the present time 
using a method of cueing that adds very greatly to the fire hazard. 
That is the use of tinfoil, or painting the edge of the film with a 
conductive paint. In one scheme, one of the rollers of the lower 
sprocket is insulated from the projector head. One wire of a 110- 
volt supply leads to the projector head ; another from the insulated 
roller leads to a lamp; thence to the other side of the 110- volt supply. 
The tinfoil or conductive paint completes the circuit from the one 
roller to the other, thus causing the lamp to light. Under normal 
conditions a very small current flows through the circuit. But, sup- 
pose the lamp were injured or something happened that would short- 
circuit the lamp as the conductive strip passed through. There would 
be a flash, and then a fire. True, there are regulations against such 
things, but they do not seem sufficient to stop the practice. How- 
ever, there are ways of stopping people from doing things, and usually 
the best way is to show them something better. A few examples of 
the fore-mentioned methods of individual cueing, now in general 
use, are illustrated in the accompanying photographs of pieces of 
film (Figs. 1 and 2) cut from prints received from exchanges. 

Requirements of a Satisfactory Change-Over Method. In order that 
a change-over may be perfect several precautions must be ob- 
served : (1) the film in the oncoming projector must be threaded at the 
proper distance from the first picture; (2) the motor switch and 

D E F 

FIG. 1. (A, B, C) Illustrating complete disregard for the standard 
dots. The edge of sample A has been painted with a conductive paint. 

(D) Since the crosses on the left are more upright than those on the 
right, the indication is that they were put on the film in different theaters. 

() The black squares represent pieces of black film cemented to the 
print. As a result, the film was made very stiff and tended to break 
more easily. 

(F) The presence of the round hole at the bottom as compared with 
the oval holes above would indicate that the holes were punched in 
different theaters. The scratches in this color print extended nearly to 
the center of the frame, and showed red on the screen. 



douser switch must be closed at the proper time; and (3) the sound 
must be changed from one machine to the other at the proper tim.>. 

The first operation can with a little care on the part of the pro- 
jectionist be done correctly. However, as to the others, 
the most alert and conscientious projectionist can not repeat them 
at the same intervals every time, and variations in the performance 
of those operations result in imperfect change-overs. From such 
facts it is apparent that change-overs can not be made perfectly 

FIG. 2. Miscellaneous examples of "cues" clipped from current films. 

every time by manual means: they must be accomplished by auto- 
matic means. It is the writer's opinion that the apparatus described 
herein fulfills the requirements for a means for making a perfect 
automatic change-over every time in any theater. 

In any remedy for the appalling situation that now exists several 
facts must be borne in mind : 

(1) If individual visual cueing is to be eliminated, all visual cues must be 

(2) The method must be amenable to standardization, and adapted to the 
needs and requirements of all theaters. 

(3) Cues or means of control of an automatic change-over system have no 
place in the picture portion of the film. 

(4) The cue or control of the device must be impossible of duplication by the 

(5) The cue or control must be positive in operation and impossible of acci- 
dental duplication. 

(6) The cue or control must be photographically recorded on the film, and 
not controlled by trimming or notching the edge of the film. 

The constant frequency area will fulfill all these requirements; 
a portion of a film thus prepared is illustrated in Fig. 3. 

The Change-Over Control. Following the trend of many modern 
electrical developments, the change-over device depends upon a 
beam of light and the photoelectric cell for its operation. Referring 
to Fig. 4, the optical assembly collects light from the exciting lamp 



[J. S. M. P. E. 

FIG. 3. A portion of film bearing the constant frequency record on its edge. 

and projects a line of light upon the film. Due to the fact that the 
frequency recorded on the film will not be high, it will be apparent 
that a rather inferior optical system, compared with that used for 
sound-on-film, may be used. 

FIG. 4. Wiring diagram, illustrating how the various circuits are 
controlled in sequence by the revolving drum. 

The photoelectric cell is coupled to the amplifier in the conventional 
manner. Current fluctuations in the cell cause fluctuations in the 
voltage drop across the resistor, which are transmitted through the 
coupling condenser to the grid of an amplifier tube. The transformer 
in the output circuit of the tube is coupled to a tuned relay, which is 
tuned to the frequency recorded on the film. 

Mar., 1934] 



We shall assume that the tuned relay has just been actuated by the 
passage of the constant frequency area through the light beam. 
Being a momentary contact relay, it closes a circuit to an interlocking 
relay. The latter closes a circuit to the motor, which is geared to a 
revolving drum in such a manner as to cause the drum to revolve 
slowly and at a predetermined speed. On the drum are means for clos- 
ing various circuits in sequence and with a predetermined interval of 
time, depending upon the adjustments of the circuit-closing means. 

The drum, now revolving, first closes the circuit of a bell or 
other warning signal, which calls the projectionist's attention to the 
fact that the reel is nearing conclusion and that it is time for him 
to light the lamp of the oncoming projector and make any other 
adjustments that may be necessary. The next circuit to close is 
that of the projector motor, which is completed through several 
switches, the purpose of which will be explained later. Next, the 
douser control relay and the sound control relay operate simul- 
taneously. The douser relay closes a circuit from a source of electric 

FIG. 5. Wiring diagram, illustrating how the various circuits may be 
controlled by separate areas of different frequencies. 

supply to the coils of the dousers (the battery shown as the source of 
supply was included as a matter of convenience in illustration only). 
The sound control relay is a momentary contact relay. Its function 


is to break the interlock of one of the coils of the double-coil inter- 
locking relay, which breaks the sound circuit and closes the circuit 
of the opposite coil of that relay, thus causing the sound circuit to be 
closed for the other machine. The change-over has thus been made ; 
the revolving drum continues to run until the circuit of the inter- 
locking relay controlling the motor is broken by the drum. The out- 
going machine is then stopped by manually operating switch 172 
(Fig. 4), which breaks the interlocking circuit of the projector motor 
relay. Due to the fact that there is no particular advantage in stop- 
ping the outgoing projector automatically, means have not been 
included on the drum for doing so. It will be understood, however, 
that any operation associated with change-overs, such as the control 
of title curtains or lighting effects, wherein the desired effect depends 
upon executing the operations at the proper instant, can be controlled 
by the drum. The drum can be made in sections, so that any number 
of sections may be used; and each section may be so adjusted in 
relation to the others so as to conform to the individual requirements 
of each theater. 

Should it be desired to change over the sound from one machine 
to the other manually, switch 169 (Fig. 4) may be used. Switch 158 
is provided for stopping the sound reproduction of either machine. 
Switch 168 is used for closing the circuit of the drum motor relay 
manually when starting the first projector at the beginning of a show, 
thus insuring that everything is operating properly and that the 
douser is not opened until the film has reached the proper position, 
eliminating the chance of showing leader footage marks on the 
screen. Switch 170 may be used when shutting down after the last 
reel of a performance, thus preventing the projector motor from 
starting. Switch 171 is used for starting the projector motor alone, 
for "warming up" or testing, or for playing musical disk selections 
on certain types of equipment. 

Fig. 5 illustrates a slight variation of the system shown in Fig. 4. 
Here it is intended that separate constant-frequency areas, of differ- 
ent frequencies, located on the film in their respective positions, will 
actuate the different relays controlling the several circuits in the 
proper order and at the proper time by varying the inductance and 
the capacitance of the separate resonance units. However, in the 
opinion of the author the revolving drum method is by far the most 
practicable, because it can be adjusted to conform to any require 
ments of each theater. 



Summary. The control frequency principle, consisting in the impression of a 
supra- or infra-audible frequency on a disk record for the purpose of actuating re- 
lays and local circuits in order to change views projected by a stereopticon or other 
means, is described. Two forms of apparatus are presented, the Phonopticon and 
the Controlophone. 

We realized some time ago that it might be desirable to synchronize 
certain mechanical and electrical actions with recorded sound, if such 
an accomplishment could be achieved by fairly simple and not too ex- 
pensive equipment; and believed that it might be done by superim- 
posing at the proper point on the record some frequency, preferably 
outside the audible range, and arranging the circuits so that this fre- 
quency would operate a relay while the remainder of the recorded 
sound operated loud speakers in the usual manner. The frequency 
might be supra- or infra-audible, but the latter was chosen. We 
planned to use ordinary lateral-cut records for the commercial equip- 
ment, on which it would be difficult to impress a high frequency, and 
from which it would be as difficult to reproduce it. Furthermore, 
the high-frequency impressions wear off more rapidly than the low. 

We therefore decided to use a low control frequency, keeping in 
mind the danger of overcutting, the poor response of the average com- 
mercial pick-up, and the possibility of interference by the very low notes 
of musical instruments. After considerable investigation, we found 
that a 48-cycle note, if recorded and reproduced as nearly as possible 
in sinusoidal form, was practically inaudible to the average person. 
This frequency was also within the other limitations, and has been 
found in practice to be very satisfactory. It is the frequency of G 3 
in the musical scale. While several instruments go below G 2 , the im- 
pression of the fundamental is not sufficient to operate the control 

We were forced to develop the Controlophone from standard com- 

* Presented at the Fall, 1933, Meeting at Chicago, 111. 
** Jenkins & Adair, Inc., Chicago, 111. 



J. E. JENKINS AND S. E. ADAIR [j. s. M. P. E. 

mercial parts, for the reason that it was necessary to make the sale 
price as low as possible. It was not possible to use special pick-ups, 
turntables, or tubes, and we were limited to the regular lateral-cut 
record. Outside of special transformers and retards, which are part of 
our business, all parts had to be purchased on the open market. 
This fact did not in any way simplify matters. 

While the apparatus was still in an early stage of design, we en- 
countered an interesting problem in connection with the 48-cycle 
oscillator for putting the impulses into the recordings. These im- 
pressions became known as "bloops," and the device was therefore 


FIG. 1. Schematic circuit of apparatus generating 48-cycle control 
frequency, used in producing Controlophone records. 

the "blooper." It had to be portable, self-contained, produce a really 
sinusoidal current of constant amplitude, connect into a microphone 
mixing position, and impress in the wax a bloop that was free from a 
most annoying click at its beginning and end. The final device (Fig. 1) 
consisted of a good tuning fork, one of its tines driven by a vacuum 
tube with grid and plate magnetically coupled, and the other tine 
affecting a simple magnetic pick-up, which was followed by a one- 
stage amplifier having a 200-ohm output. A remotely controlled re- 
lay was included in the circuit, as well as a section of low-pass filter. 
The purpose of the filter was to annul the objectionable click caused 
by transients that occurred when the relay happened to close the out- 

Mar. 1934] 



put circuit at the instant of maximum amplitude of the 48-cycle cur- 
rent. For reasons that will be noted later, constant frequency was 
not a requirement. A variation of 3 per cent would do no harm; 
hence, thermal control and like complications were unnecessary. 

The Phonopticon and Controlophone circuits are basically identical 
(Fig 2). The Phonopticon was designed to work in conjunction with 
the Bausch & Lomb Balopticon, a continuous automatic slide projec- 
tor, motor-driven. It may be equipped with a limit switch, which 
will stop the motor after a new slide has been moved into position. 
This limit switch is simply a cam-operated contact in series with the 








46 TO 50 CYCLES 


FIG. 2. Schematic circuit of Phonopticon. 

motor circuit. The function of the Phonopticon is to short-circuit 
the switch when the pulse on the record closes a simple relay. The 
pulse on these records has a duration of l 1 /^ seconds. The new slide 
comes up about half a second after the start of the pulse. The 
Balopticon mechanism cycles once, and awaits the next pulse. The 
result is an inanimate automatic lecturer, which delivers a talk on a 
given subject and turns up its own slides at exactly the right moment. 
The turntable is of the automatic repeating type, so that the machine 
may be run continuously. If desired, the turntable is equipped with 
a limit switch, so that the Phonopticon will tell its story once, and 



[J. S. M. P. E. 

then wait until the starting button is pressed again. The length of the 
record, when concerned with sales talk, should never exceed three or 
four minutes. Lectures, of course, may be of any length. In one 
instance, we had to supply a turntable with a record changer, to pro- 
vide a performance lasting 40 minutes, and using 80 slides. This is 
not desirable. 

While up to the present time our installations have included a 
Balopticon, the device is applicable to any automatic picture-changing 
mechanism. It is now being arranged for use with a slide-film pro- 
jector in which the film will be moved by a solenoid, this latter being 
operated by the pulses on the record. 

The applications of the Controlophone are far more difficult to de- 
scribe, as they have assumed such a number of different forms. As 



FIG. 3. Relay and switching circuits of Controlophone. 

has been said, the basic set-up is the same, as shown in Fig. 3. Follow- 
ing the sensitive d-c. relay in the plate circuit of the control output 
tubes is a time-delay relay. It is set to delay about 2 / 5 of a second, the 
purpose being to prevent its acting on brief impulses from scratches or 
imperfections in the record and thus throwing the system out of syn- 
chronization. The contacts of the time-delay relay close the motor 
circuit of a standard rotary selector switch, which in turn controls the 
lights, motors, etc., which go to complete the Controlophone display. 
Between the selector and these devices are interposed suitable exter- 
nal relays, the size and type of which are governed by the loads 

The selector switch is wired so that it returns to its starting position 
at the end of each performance. Instead of depending upon the 

Mar., 1934] 



pulses in the record to do this, we decided on a definite mechanical 
control, and built into the repeating turntables a contactor that oper- 
ates only during the repeating action. The circuit established 
through this contact restores the switch, and re-synchronizes the sys- 
tem, if it be necessary to do so. 

A brief description of some Controlophone installations should be 
of interest, and I have selected three which are very different in 

A simple application is found in an installation demonstrating a 
mayonnaise mixing machine. The record runs 3 l /z minutes, during 

FIG. 4. Standard Oil Co. of Indiana display in operation. 

which time the announcer describes the mixing machine. As he men- 
tions the various important parts, the attention of the audience is at 
once directed to them by neon tube arrows, suspended over them and 
pointing to them. 

As another example, part of the Union Carbon & Carbide exhibit 
in the Hall of Science,* is actually a large 12-foot turn-table support- 
ing four dioramas pertaining to various products, entirely housed in, 
and having three apertures through which three of the dioramas may 
be seen. Four different records are required, and four voice channels. 

* Century of Progress World's Fair, Chicago, 111. 



[J. S. M. P. E. 

Only one of these is a Controlophone, which supplies the oral descrip- 
tion and operates dimmers and various lights on the acetylene 
diorama. The other three dioramas do not require synchronous 
effects, but their three records are started by the Controlophone 
simultaneously with its own record. All four records are of the same 
duration, and when their speeches have ended, the Controlophone 
rotates the whole 12-foot turntable one-quarter of a revolution, after 
which it starts all four shows over again. People grouped around the 
structure may see the four shows in succession without changing their 
positions. The four records contain 2 1 / 2 -rnmute talks. 

The largest and most complex installation is for the Standard Oil 
Co. of Indiana, in the huge rotunda of the Travel and Transport 

FIG. 5. Standard Oil Co. of Indiana control room. 

Building* (Fig. 4). The 32-ton steel structure in the center of the 
rotunda is about 90 feet high. The three lower stages house sil- 
houette machinery, with three 15-kw. lamps centrally located. The 
ring above these supports 24 moving-coil loud speakers. The crown 
contains four silent 35-mm. projectors and operators. 

The show begins with a 6-minute introduction, during which a 
musical prelude, composed for the purpose is reproduced. Pulses on 

* Century of Progress World's Fair, Chicago, 111. 


the record control the silhouette machines in synchronism with the 
changes in mode and tempo of the musical score, slowly building the 
lighting and sound up to a very startling climax. At that point, the 
projectors are started, and the Controlophone continues to supply 
suitable background music for the moving pictures until their con- 
clusion, some 20 minutes later. The show is started by a time-clock, 
hourly on the half -hour. The only manual operation is the reloading 
of the projectors by the operators. 

The control room, shown in Fig. 5, contains, from left to right, 
generators for the projector arcs and for the Controlophone, which is 
seen mounted on three tall racks ; then the complicated relay panels, 
and under them the generators for the silhouette lamps; and finally 
the program drum, the rotation of which is handled directly from the 
rotary switch of the Controlophone. 



Modern Motion Picture production has developed an interesting technic 
which, though not entirely new, is now applied to such an extent that it has 
required the development of new apparatus to assure perfection of execution 
and rapidity of manipulation. In 1910 or 1911, the Italian producer of Cabiria 

FIG. 1. The Rotambulator, show- 
ing highest position. 

conceived the idea of replacing the stereotyped system of cutting from long shots 
to close-ups with the perambulating of the camera toward the action that was 
to be emphasized. 

All who saw the picture marveled at it then, and still remember today the 

* Presented at the Fall, 1933, Meeting at Chicago, 111. 
** Bell & Howell Co., Hollywood, Calif. 


theretofore unknown sense of continuity of action that resulted and the sense of 
intimacy that was conveyed by that relatively simple trick. It was only around 
1926 that German producers revived the idea and elaborated it with such as- 
tounding results that American producers fell in line and the camera began to 
travel about the sets, at times, perhaps, with exaggeration, but most of the time 
maintaining a very effective continuity of action and a sustained interest in the 

At the time that recorded speech became an essential part of motion pictures, 
the new technic proved invaluable as a means of sustaining the tempo of the 
action which the spoken word had a tendency to slacken. Expedients were 
resorted to and the camera was set on rude perambulating platforms improvised 

FIG. 2. The elevating mechanism. 

according to the needs of each individual scene. Little by little, with the de- 
signing of more efficient rolling tripods, the perambulating camera had to evolve 
means for controlling the technic involved in its motion, with the result that 
special "follow focus" devices and special self-adjusting finders were added to it. 
The increasing complexity of camera motion brought forth the building of 
cranes, some of them mastodonic, some of less bulk and more easily manipulated. 
At that time the Bell & Howell Company considered the advisability of con- 
structing a camera stand for practical every-day use one that would permit 
with ease the simultaneous use of the four elements of camera motion. These are, 
disregarding any attempt of having the camera perform acrobatic tricks: per- 
ambulating, panning, tilting, and elevating or lowering. Such a piece of equip- 



[J. S. M. P. E. 

ment was considered highly desirable also because the bulk and weight of the 
camera "blimps," now in use, made every camera "set-up" a matter of brawn 
rather than brain unless facilities were given the cinematographer to set his 
camera easily and quickly at the proper distance and height. 

The Rotambulator , illustrated in Figs. 1 and 5, consists of a three-wheeled 
undercarriage on which rests a rotating platform. A strong upright holds the 
camera platform and the elevating and tilting devices. Both panning and tilting 
are accomplished by the camerman from a seat which is an integral part of the 
panning platform, so that his position in relation to the camera is always the same 
irrespective of panning and perambulating. 

FIG. 3. 

The base, showing the large pulley for the panning drive 
and the large ball race. 

Fig. 1 shows the general appearance of the apparatus with the camera plat- 
form at its highest level. The main dimensions are as follows: 

Over-all height 

Over-all length 

Over-all width 

Max. height of camera table 

Min. height of camera table with panning wheel attached 

Min. height of camera table with panning wheel removed 

Diameter of rotary platform 

Size of camera table 

Max. height of seat 

Min. height of seat 

Height of standing platform 

Net weight 

90 in. (7 ft. 6 in.) 
62 3 /4 in. (5 ft. 2 3 A in.) 
46 3 /4in. (3 ft. 10 3 /4in.) 
71 3 /4in. (5ft. H 3 / 4 m.) 
16 in. (1ft. 4 in.) 
12 3 / 4 in. (1ft. 3 / 4 in.) 
42 in. (3 ft. 6 in.) 
13V4in. X 13 7 /shi. 
39 3 / 4 in. (3it.3 3 / 4 in.) 
31 3 /4 in. (2 ft. 7 3 / 4 in.) 
21 in. (1ft. 9 in.) 
700 pounds 

When the camera is set at such height that it is difficult for the operator to 
follow the action through the finder, the seat is replaced by a lower platform on 
which he can stand, bringing his eyes to the level of the finder. 

The elevating mechanism is illustrated in Fig. 2. Through a crank and pulleys 


linked by a belt, the elevating screw is made to rotate at a ratio of five turns 
for three complete turns of the crank. The screw acts upon a nut in the camera 
platform housing, raising, or lowering it one foot for every 36 turns of the crank. 

A great deal of thought has been given to determining the elevating speed, 
experience having taught that it would be mostly used for rapidly setting the 
camera at the proper height and seldom for the purpose of achieving the effect 
of changing elevation while the camera is operating. That can, however, be 
done, and again experience has proved that a greater or lesser elevating or de- 
scending speed would not be conducive to any better results. 

The panning crank is in easy reach of the operator at the left of the camera 
platform. It operates through two beveled gears which rotate an upright and a 

FIG. 4. The camera platform and the panning and tilting 

pulley located within the base which is in turn linked by a belt to another pulley, 
shown in Fig. 3. The smoothness of the panning operation is attained through 
proper tension of the belt and the possibility of adjusting it, and through a large 
ball race as shown in Fig. 3. 

The tilting mechanism is easily controlled by the right hand of the operator, 
and consists of a worm and worm gear which control the motion of a system of 
pulleys and an adjustable belt shown in Fig. 4. Ball races of generous diameter 
assure smoothness of motion and a good balance against the considerable weight 
of a camera enclosed in a blimp (from 300 to 500 pounds). The camera platform 
is a separate unit solidly anchored to the tilting system. Its design can be altered 
to accommodate blimps of any design. 

Both the panning and tilting devices can be disengaged; the first by releasing 
a clutch ; the second by opening a lock, thus idling the gears so that the apparatus 



[J. S. M. P. E. 

can be used in the same manner as a "free tripod head," in which case, however, 
the operator must work from the floor and sacrifice the convenience of the seat. 

The undercarriage forms a solid triangular base of such width and length as 
to assure rigidity of the apparatus even when the camera is at the maximum 
height. The carriage is mounted on four rubber rimmed wheels mounted on ball 
bearings and perfectly aligned to insure smoothness of running. An operator 
perambulates the apparatus by means of a tongue handle, the two wheels at the 
apex of the triangle being mounted on a swivel carriage. Three jacks permit 

FIG. 5. 

The Rotambulator in use at the M-G-M 

stabilization of the machine for stationary "shots" and also leveling in case of 
slight unevenness of the floor. 

Among the details of construction, the following are worthy of mention: 
the camera platform is so located that in the panning operation the axis of revolu- 
tion is as close as possible to the photographic lens, depending upon the design of 
the blimp. The cameraman's seat is adjustable so that the operator can place it in 
the most comfortable position. Sockets are made part of the rotating platform 
so that additional seats or lighting equipment can be fastened to the apparatus. 

The other type of Rotambulator is illustrated in Fig. 5, where it is shown with 
the camera blimp in position, and cameraman William Daniels and Director 


Edgar Selwyn at the controls. The principal difference between this design and 
the one described above consists in the method of controlling the tilting and 
panning arrangements. Gears, pulleys, and belts are eliminated, smoothness 
of operation being attained by controlling the mechanism through oil feeds. The 
tilting handle is simply moved up or down according to the requirements, and a 
slight pressure on a trigger which is integral part of the handle releases a stout 
brake which otherwise holds the tilting device locked in position. 

The panning arrangement is novel in that it is operated by a slight pressure 
of the operator's foot on a stationary circular platform independent of the ro- 
tating platform on which the camera standard is mounted. This method offers 
the advantage of freeing the left hand of the operator. 


Commercial Cinematography. G. H. SEWELL. H. Greenwood & Co., London, 

There has been a need for a book of this type which would give the industrial 
firms definite practical information concerning the making of 16-mm. pictures. 
As the author states, "The making of films consists of one part photography and 
nine parts picture making." Too often in his experience business firms have 
overemphasized the photographic aspect and have given too little thought to 
the idea of producing an interest-compelling picture. Equipment is described 
for taking, editing, and projecting 16-mm. and 9.5-mm. films. Data are included 
on stop-motion and cartoon work, and a brief concluding chapter deals with 
amateur sound films. 


Amateur Talking Pictures and Recording. BERNARD BROWN. Isaac Pitman 
& Sons, London and New York, 1933. 225 pp. 91 figures. 

The author states in his preface, "The ordinary gramaphone is out of date, 
radio is almost commonplace, and television is somewhat in the future. The home 
talking picture might well fill the gap." This terse statement describes clearly 
the reason for writing the book; and this little volume provides the amateur 
with an excellent summary of the available equipment for small-scale sound 
recording as well as working details for actually doing it. Sound-on-disk and 
sound-on-film methods are described. The book does not aim to include every 
type of amateur sound equipment available but is intended to cover those that 
differ essentially in fundamental design. The illustrations are well chosen, and 
the drawings assist materially in clarifying the text. 


The Sound Motion Picture in Science Teaching. PHILLIP J. RULON. Har- 
vard Univ. Press, Cambridge, Mass., 1933. 236 pp. 

There has been considerable divergence of opinion relative to the value of 
motion pictures as a visual aid in educational work. In the introduction, the 
author states ". . .only a small percentage of the literature appearing during the 
last decade has concerned itself with experimental evidence on the effectiveness 
of such aids." There are so many easily overlooked factors that exert an im- 
portant influence on the results of such work that the author feels that even in 
those cases where experimental investigation has been attempted, "... the motion 
picture as an instructional device is yet to be evaluated." 

This book represents a report of ". . .an attempt to evaluate numerically the 
educational effectiveness of the sound motion picture in the teaching of a school 
subject." The ninth grade (first year high school) was used, and the subject 
chosen was General Science, particularly Physiography and Biology. A text- 
book was prepared that was designed to be typical of those in the fields of general 
science, and films were specially produced to parallel the text. Of two groups 


of pupils, one used the text-book only; the other the text and films. A third 
group, serving as a second control group, did not study the experimental in- 
structional material. The children were drawn from three suburbs of Boston. 
The school year was divided into three parts: (1) a pre-instructional period, 
(2) the experimental instructional period, and (3) a retention period when the 
regular general science work was taken up again. Tests were given at the end of 
each period. 

The general conclusions of the results of the work were (1) the general pupil- 
achievement increase ascribable to the use of the film exceeded 20 per cent; 
(2) those facts and relations specifically dealt with in the film resulted in a 35 
per cent increase in pupil achievement; and (3) neither of the gains mentioned 
under (1) and (2) were made ". . .at the expense of more important but less 
definable educational values, such as good habits of thinking." 

The first half of the book reviews the details of the experiment, and the latter 
portion contains a bibliography, a census of occupational listings, scripts for experi- 
mental films, bulletins, teachers' guides, and data used in the tests. 

G. E. Matthews 



At the last meeting of the Board of Governors, held on January 19th at New 
York, final steps were taken to put into effect the amendments of the Constitution 
and By-Laws proposed at Chicago, October 16th, and approved by letter ballot 
of the Active membership of the Society January 15, 1934. The Constitution 
and By-Laws now in effect are as embodied in the brochure mailed with the 
voting ballots to the Active membership the latter part of October. 

The system now in effect is, in brief, as follows : the Board of Governors con- 
sists of the President, the Past-President, five Vice- Presidents, the Secretary, 
the Treasurer, five elective governors, and the three section chairmen. The 
nominations made by the Board of Governors were as follows: 

Executive Vice-President, H. C. Silent 
Engineering Vice-President, L. A. Jones 
Editorial Vice-President, J. I. Crabtree 
Financial Vice-President, O. M. Glunt 
Convention Vice-President, W. C. Kunzmann 
Governor, A. S. Dickinson 

Under the new plan the various Committees of the Society will be under the 
direction of the several Vice- Presidents, viz., Engineering Vice-President: all 
the technical Committees; Editorial Vice-President: the Board of Editors, 
Historical Committee, Progress Committee, and Papers Committee; Financial 
Vice-President: the Ways and Means Committee and the Membership Com- 
mittee; Convention Vice-President: the Convention Arrangements Committee 
and the Publicity Committee. 

Ballots for voting on the nominations have been mailed to the Active member- 
ship, and will be counted on April 19th, whereupon the elected nominees will 
immediately assume office. The remainder of the Board of Governors, with the 
exception of the two incumbent Vice- Presidents, will be as presented on the 
reverse of the contents page of this issue of the JOURNAL. 


At the same meeting of the Board, final steps were taken to put into effect the 
new annual dues rates, and to establish the new three grades of membership that 
will replace the existing two grades. The new grades and rates are in the table 
on the next page. 

The fiscal year of the Society will now coincide with the calendar year, be- 
ginning January 1st, instead of October 1st. On that account, an adjustment 
of the dues of the current year was necessary in order to encompass the year and 



a quarter from October 1, 1933, to January 1, 1934. Accordingly, all members 
were billed for the last quarter of 1933 at the old rates (*. e., $5 for Actives, $2.50 
for Associates) to which were added the normal annual rates given in the table 
below. Thus, the dues of Active members for the year and a quarter were $15, 
and for Associate members, $8.50. Those members who paid their dues in full 
according to the old rates (Active, $20, Associate $10) will be credited with the 
excess against their dues for 1935; i. e. t Active members will be credited $5, and 
Associate members $1.50. 



Years of 
M. P. 

No. of 






3 Fellows 






3 Fellows or Actives 

$ 5 




1 Fellow or Active 

$ 6 

* References should be named who have personal knowledge of the applicant's experience. 
It is suggested that applicants furnish more than the requisite number of references and a full 
record of experience. 

** First year's dues prorated monthly from date of admission; dues of succeeding years 
payable January 1st. 


A Committee was appointed by the Board of Governors for the purpose of re- 
classifying the membership of the Society into the three new grades. Invitations 
to apply for transfer to an upper grade are being extended by the Committee to 
those members whose records on file indicate their fitness for the grades in question. 
All members, of course, are entitled to apply for transfer voluntarily, if they so 
wish: the approval of their application will depend, in the usual manner, upon 
their ability to comply with the qualifications and requirements specified in the 
By-Laws (By-Law I, Sec. 1). The Committee is composed of the following 
members : 

T. E. SHEA, Chairman 



1934 BUDGET 

Two budgets were approved by the Board of Governors: one to terminate the 
accounts of the Society on January 1, 1934, in order to advance the date of com- 
mencement of the fiscal year; and another to encompass the calendar year 1934, 
which henceforth will be identical with the fiscal year. Under the new arrange- 
ments the Society will operate within its income; and, if the campaign for new 
members proves as successful as is anticipated, may be able to broaden its activities 
and improve its JOURNAL considerably. 


The Board of Governors ruled that the names of the three local sections of the 
Society, now known as the New York Section, the Chicago Section, and the 
Pacific Coast Section, shall hereafter be the "Atlantic Coast Section," the "Mid- 
West Section," and the "Pacific Coast Section," respectively. 



W. C. KUNZMANN, Chairman 



H. BLUMBERG, Chairman 





H. GRIFFIN, Chairman 

Officer and Members of Atlantic City Local No. 310, I. A. T. S. E. 


MRS. M. C. BATSEL, Hostess 

Assisted by 




The Convention will convene at 10:00 A.M., Monday, April 23rd, at the Chal- 
fonte-Haddon Hall, in the Viking Room on the thirteenth floor of the Haddon 
Hall section. At noon of the opening day there will be an informal get-together 
luncheon, during which the members of the Society will be addressed by several 
prominent speakers. The morning preceding the luncheon will be devoted to 
registration, reports of officers, and other Society business, as well as the reports 
of technical committees. 


All technical sessions and film exhibitions will be held in the Viking Room, 
where also will be located the registration headquarters. Technical sessions 
will be held on Monday, Tuesday, and Thursday afternoons, and on Tuesday, 


Wednesday, and Thursday mornings. Monday morning will be devoted to 
Society business and committee reports; Wednesday afternoon, preceding the 
semi-annual banquet in the evening, will be left free for recreation. The film 
programs of recently produced outstanding features and shorts will be held on 
Monday and Tuesday evenings, and will be booked by Mr. J. Greenburg, of 
the Philadelphia Film Board of Trade, and Mr. H. Blumberg, chairman of the 
Local Arrangements Committee. 


The S. M. P. E. Semi-Annual Banquet and Dance will be held in the Vernon 
Room of the Chalfonte-Haddon Hall on Wednesday, April 25th, at 7:30 P.M. 
an evening of dancing, movies, and entertainment; no banquet speeches. Ban- 
quet tickets should be obtained at the registration headquarters : tables reserved 
for six or eight persons. 


Excellent accommodations are assured by the management of the hotel, and 
minimum rates are guaranteed. Room reservation cards mailed to the member- 
ship of the Society should be returned immediately to the Chalfonte-Haddon 
Hall in order to be assured of satisfactory reservations. 


Room with bath, ocean view, single $4.00 
Room with bath, ocean view, double $6.00 
Room with bath, city view, single $3.00 
Room with bath, city view, double $5.00 


A reception suite will be provided for the use of the ladies attending the Con- 
vention, and an attractive program for their entertainment is being prepared by 
the Ladies' Committee. 


Arrangements are being made to hold an exhibit of newly developed motion 
picture apparatus, in order to acquaint the members of the Society with the 
newly devised tools of the industry. The exhibit will not be of the same nature 
as the usual trade exhibit; there will be no booths, but each exhibitor will be 
allotted definite space and all exhibits will be arranged in a single large room. 
Requests for space should be directed to the General Office of the Society, 33 
West 42nd Street, New York, N. Y., stating the number and nature of the items 
to be exhibited. Regulations concerning the exhibit are given in the tip-on 
in this issue of the JOURNAL. The charges for space will be as follows: up to 
20 sq. ft., $10; every additional 10 sq. ft., $5. 



The last meeting of the Projection Practice Committee, held at New York on 
January 24th, was perhaps the most outstanding meeting of the Committee since 
its organization. Twenty-one representatives of the major theater circuits 
attended, as also representatives of the National Carbon Company, for the 
purpose of discussing and analyzing the various problems attendant upon the 
new a-c. carbon arc projector lamp. The Committee plans to include, as part of 
its semi-annual report, a full presentation of the operating features of the lamp 
from the projectionist's point of view. The next meeting of the Committee will 
be held on February 28th. 


On January 30th, the members of the Atlantic Coast Section were the re- 
cipients of a kind invitation of the Bell Telephone Laboratories to attend a demon- 
stration of transmission and reproduction in auditory perspective by Dr. Harvey 
Fletcher, at the Engineering Societies Building, New York, N. Y. Full details 
of the demonstration are being prepared by Dr. Fletcher, and will be published in 
the JOURNAL in the near future. The meeting was attended by approximately 
700 members of the S. M. P. E. and the Acoustical Society of America. 

The next meeting of the Atlantic Coast Section will be held on April 28th, at 
New York, N. Y. 




Volume XXII APRIL, 1934 Number 4 



Equipment for Recording and Reproducing Sound with Photo- 
Film (Concluded) A. F. CHORINE 215 

Acoustical Requirements for Wide-Range Reproduction of 
Sound S. K. WOLF 242 

Wide-Range Recording F. L. HOPPER 253 

The "Selenophon" Sound Recording and Reproducing System. . 

G. E. ROTH 260 

Society Announcements 270 

Spring Convention: Atlantic City, N. J., April 23-26, 1934 273 





Board of Editors 
J. I. CRABTREE, Chairman 



Subscription to non-members, $8.00 per annum; to members, $5.00 per annum, 
included in their annual membership dues; single copies, $1.00. A discount 
on subscriptions or single copies of 15 per cent is allowed to accredited agencies. 
Order from the Society of Motion Picture Engineers, Inc., 20th and Northampton 
Sts., Easton, Pa., or 33 W. 42nd St., New York, N. Y. 

Published monthly at Easton, Pa., by the Society of Motion Picture Engineers. 

Publication Office, 20th & Northampton Sts., Easton, Pa. 
General and Editorial Office, 33 West 42nd St., New York, N. Y. 
Entered as second class matter January 15, 1930, at the Post Office at Easton, 
Pa., under the Act of March 3, 1879. Copyrighted, 1934, by the Society of 
Motion Picture Engineers, Inc. 

Papers appearing in this Journal may be reprinted, abstracted, or abridged 
provided credit is given to the Journal of the Society of Motion Picture Engineers 
and to the author, or authors, of the papers in question. The Society is not re- 
sponsible for statements made by authors. 

Officers of the Society 

President: ALFRED N. GOLDSMITH, 444 Madison Ave., New York, N. Y. 
Executive Vice-President: HAROLD C. SILENT, 7046 Hollywood Blvd., Los 

Angeles, Calif. 

Engineering Vice-P resident: LOYD A. JONES, Kodak Park, Rochester, N. Y. 
Editorial Vice-President: JOHN I. CRABTREE, Kodak Park, Rochester, N. Y. 
Financial Vice-President: OMER M. GLUNT, 463 West St., New York, N. Y. 
Convention Vice-President: WILLIAM C. KUNZMANN, Box 400, Cleveland, Ohio. 
Secretary: JOHN H. KURLANDER, 2 Clearfield Ave., Bloomfield, N. J. 
Treasurer: TIMOTHY E. SHEA, 463 West St., New York, N. Y. 


EUGENE COUR, 1029 S. Wabash Ave., Chicago, 111. 
HERFORD T. COWLING, 7510 N. Ashland Ave., Chicago, 111. 
ARTHUR S. DICKINSON, 28 W. 44th St., New York, N. Y. 
RALPH E. FARNHAM, Nela Park, Cleveland, Ohio. 
HERBERT GRIFFIN, 90 Gold St., New York, N. Y. 
EMERY HUSE, 6706 Santa Monica Blvd., Hollywood, Calif. 
WILBUR B. RAYTON, 635 St. Paul St., Rochester, N. Y. 
HOMER G. TASKER, 41-39 38th St., Long Island City, N. Y. 


(Concluded from page 172 of the March Journal) 

Summary The work of the Central Laboratory of the All- Union Electrical 
Trust, Leningrad, U. S. S. R., under the direction of the author, in connection with 
the study, design, and manufacture of motion picture recording and reproducing ap- 
paratus is described. The present installment of the article deals with the research 
on light modulators, varible width and variable density; ordinary and noiseless 
recording systems; various forms of recording light sources; and the relative char- 
acteristics of the different kinds of modulators, viz.. Western Electric, RCA, the 
author's, Kerr's cell, and flashing lamp. 


In our work it was necessary to produce equipment of the simplest 
construction possible, so that repairs could be made at the existing 
technical shops of the picture studios themselves. The apparatus 
delivered for practical use can be classed according to the following 
categories : 

(1) Stationary apparatus for studio use, recording sound on a positive film 
separately from the picture. 

(2) Light, portable apparatus for field use, recording sound and picture on the 
same negative film. 

(3) Light apparatus for newsreel recording on a single film. 

(4) Universal apparatus that can be used in either the studio or the field. 

(5) Stationary apparatus for recording sound on 300-meter lengths of film. 

The author's modulator has been used in all such apparatus except 
in apparatus of the third class, where recording was done by a flashing 

Fig. 20 shows the earliest model of light modulator, and camera 
for 120 meters of film, actually delivered for use. The magnet is of 

* Received October 2, 1933. Limitations of space have necessitated abbrevia- 
tion of the original paper. 

**Director, Central Laboratory, All- Union Electrical Trust, Leningrad, 
U. S. S. R. 




[J. S. M. P. E. 

horseshoe type, with magnetizing coils wound cylindrically around it, 
and the axis of the optical system has a bend between the lamp and 
the magnetic system. With this model it was not possible to make 
variable- width and variable-density records alternatively. In all 
succeeding models the axis of the ribbon could be given any 
direction in the vertical plane by turning the entire system about 
the horizontal axis, which is the optical axis of the rays passing 
through the pole pieces of the electromagnet. 

FIG. 20. (Upper) Earliest model of the light modulator. 

FIG. 21. (Lower) Second model of the modulator, which could be 
used alternatively for variable-width or variable-density recording. 

Such an arrangement was achieved in the second model of the 
modulator (Fig. 21), by mounting the magnetic system on roller 
bearings on vertical columns, so that the electromagnet, the oil con- 
tainer soldered to it, and the frame with the stretched ribbon could 
be turned to any angle. The ribbon can be precisely adjusted in 
relation to the optical axis by the screw on the top of the oil con- 

In both models normal microscopic ten-fold anachromatic objec- 
tives, with aperture //0.30, were used. These greatly simplified 

April, 1934] 



the focusing of the optical system and made it possible to obtain a 
strong magnetic field of about 22,000 gauss in the interpolar space. 
With a 20x air objective, the best possible construction of the pole 
pieces lowered the magnetic intensity somewhat, but with the 20x 
oil-immersion objective the full field strength was restored. The 
duplicator is contained in a brass tube mounted in the opening in the 
pole piece that contains the recording objective. The shorter working 
distance of the 20x objectives necessitated a change in the construc- 
tion of the inner part of the pole pieces. Bringing the objective nearer 

FIG. 22. (Left) The fourth model 
of the modulator: ab and cd are springs 
that stretch the ribbon; the brass box 
serves both as oil container and as 
ribbon mount. 

FIG. 23. (Right) Schematic arrange- 
ment of recording mechanism. 

the ribbon reduced by some 25 per cent the strength of the magnetic 
field in the space between the poles. 

An integral part of the light modulator is the observing micro- 
scope, which visually assists in focusing the objective and controlling 
the amplitude of the vibrating ribbon. In the system described, it is 
mounted perpendicularly to the direction of the light rays, be- 
tween the projecting microscopic objective and the cylindrical lens, 
and consists of a flat reflecting mirror, a 1.5x objective, and a lOx 
ocular placed in a metallic tube mounted on a vertical support. The 
tube can be placed in any position in relation to the optical axis. 



[J. S. M. P. E. 

The operator is thus enabled to watch the amplitude of the ribbon 
during recording and regulate it by adjusting the amplifier. 

Because of its short focusing distance, the cylindrical lens must 
always be placed near the film and be focused very exactly. Conse- 
quently, in all modulators of this type the cylindrical lens is installed 
on the same mounting as the mechanism holding the film as it 
passes the cylindrical lens. Since the image can not be exactly 
focused on the film visually, even with the aid of a microscope, exact 
focus is achieved by placing the cylindrical lens in several different 
positions in relation to the film, making an instantaneous exposure 

FIG. 24. Portable sound recorder. 

in each position, and computing the proper position from measure- 
ments made on the developed film. 

Usually the cylindrical lens is placed, together with its diaphragm, 
in a special fixture installed in the short tube. In systems in which 
the light modulator is placed within the camera, this tube is mounted 
either on the wall of the camera, or on a special fixture holding the 
film carrying mechanism. 

As a source of light the standard incandescent lamp "GOZ" (12 
volts, 3 amperes) is used. Its support is designed to permit its 
adjustment to the correct position. The whole optical system of the 
modulator, except the cylindrical lens, is assembled on one block, and 
can thus be focused in advance by removing it from the recorder. 


When reinserted, it must merely be placed correctly in relation to the 
cylindrical lens. 

The fourth model is quite different in construction from the fore- 
going. The ribbon is mounted and stretched not on a separate frame 
but in a shallow round brass box that serves both as oil container 
and ribbon frame, with a removable round cover that seals the box 
when bolted on. The ribbon is stretched by two flat springs ab and 
cd (Fig. 22) whose tension is controlled by the screw / with two con- 
tacts, m and n. The amount of tension can be read from the gradu- 
ated head of screw S. Screw /, passing through screw S, stretches 
the ribbon until the spring ab touches the contacts m and n, and so 
extinguishes the control lamp. Since only the heads of screws S and 
t are outside the box, the ribbon can be stretched to the proper tension 
without opening the box. The pole pieces of the magnet, of the same 
form as in previous models, are soldered to the top and bottom of the 
box. When the ribbon is installed, its position in relation to the pole 
pieces and the opening 0, through which the light passes, can easily 
be seen and adjusted. 

To change the ribbon, it is necessary only to substitute a second 
reserve box in which the ribbon has already been properly stretched. 
This scheme eliminated many of the disadvantages of the former 
systems, although the objectives may have to be refocused. The 
system can be used either with or without the damping oil. In the 
former case the openings O at the bottom and the top of the box are 
sealed by flat-parallel glass plates preventing seepage of the oil 
from the box and into the objectives. 

The magnetic system is cylindrical in form, and consists of two 
removable parts, each having a winding and a pole piece carrying an 
objective. The entire modulator can be turned to make possible 
either variable- width or variable-density recording. 

Stationary models of sound recording equipment were assembled 
either on massive tables or on rolling platforms, along with the 
amplifier and all necessary sources of current for the amplifier and 
the light modulator. In these models there was no difficulty in using 
sufficiently powerful motors to drive the film, either of the syn- 
chronous or the interlocking type. The necessary synchronizing 
power of the motor, not less than 75-100 watts, made it possible to 
use a simple flywheel directly on the motor shaft, rotating at 1440 
rpm., in order to suppress irregularities of rotation. The shaft of the 
sprocket that drives the film is geared to the motor shaft by 1 :4 gears. 



[J. S. M. P. E. 

With only the flywheel to serve as filter, the uniformity of motion 
of the film has proved quite satisfactory. The use of so simple a 
filter was possible, however, only in mechanisms that have a con- 
stant or a slowly and smoothly varying load, and when the flywheels 
are very well made. To avoid sudden irregularities of load due to 
friction in the winding bobbin, the bobbin is driven through a round 
rubber belt, which smoothes out the effects -of bumping on the motor 
shaft. Recording on the film as it passes over the sprocket causes 
no serious distortion if the drum is made with sufficient exactitude 
and if an additional drum is interposed between it and the winding 
bobbin so that tension from the latter does not reach the former. 

'FiG. 25. Driving mechanism of portable sound recorder 
shown in Fig. 26. 

The scheme of the entire installation, including the film driving 
mechanism, is shown in Fig. 23. A is the synchronous motor, B 
a coupling, C the flywheel, D the shaft of the sprocket on which the 
recording is done, K the shaft of the interposed drum, F the tube 
with the cylindrical lens, and / the light modulator. 

Portable sound recording units, usually fixed on tripods, are too 
light to permit the use of sufficiently powerful driving motors, and 
since sound and picture are usually recorded simultaneously on the 
same film, the load on the motor driving both the camera and the 
film driving mechanism varies quite widely. This makes it rather 
difficult to move the film sufficiently smoothly. A portable model 

April, 1934] 



that has given good results in the field under varying conditions is 
shown in Fig. 24 and Fig. 25 shows its film driving mechanism. The 
d-c. motor A , consuming 30 to 35 watts at 24 volts and turning at 
3000 rpm., turns the shaft B at 1440 rpm., which rotates the 
mechanical filter C with the spiral spring D at 360 rpm. The rotation 
of the motor is stabilized by the electric regulator E. 

The shaft B rotates also the mechanism of the picture camera 
through an elastic spindle 6 mm. in diameter, housed in the bent 
tube F. Tube and spindle can be readily removed and replaced. 


FIG. 26. (Left) Universal recorder for single-film system. 

FIG. 27. (Right) Original model of universal recorder adapted to double- 
film recording 

The elastic spindle greatly reduces the'unevenness of load due to the 
operation of the "pull-down" of the camera. Through a slot at 
the bottom of the camera, the film passes to the take-up mechanism. 
The sound is recorded on the film as it passes over the smooth 
massive drum K, which rotates freely on its shaft and is driven by the 
film itself. The cylinder L stretches the film over the recording drum, 
and its escape over this drum is controlled by the sprocket TV mounted 
directly on the same shaft as the mechanical filter. The filter is 
undamped. The tube with the cylindrical lens and the focusing 
screws are mounted on the support T. Measuring and directing 



[J. S. M. P. E. 

instruments and the observing microscope are assembled on the side 
covers of the apparatus, as shown in Fig. 24. 

Sound recording with such equipment is satisfactory, but the uni- 
formity of motion of the film is insufficiently assured because of the 
small power of the motor and the relatively light weight of the filter, 
which has a natural frequency of 3 cycles per second. 

The universal equipment for recording sound and picture either 
together on a single film or separately on two films is similar in con- 
struction to the portable equipment, because it is necessary to use 
tripods and to be able to move the equipment about. Fig. 26 shows 
the adaptation of the apparatus for recording sound and picture on the 

FIG. 28. Stationary model of 
portable recorder. 

FIG. 29. Another view of Fig. 28. 

same film. The motor is so mounted that it can be easily replaced 
by an arrangement permitting the apparatus to be connected to the 
interlocked motor necessary when the sound is recorded on a separate 
film. In the latter case the apparatus is removed from the tripod and 
installed on a base plate on which the interlocked motor is mounted. 
The camera is removed and replaced by a light-tight box of the same 
dimensions, containing small cases of film and the winding mechanism. 
About thirty minutes are required to change from the one form of 
equipment to the other. The reliability and quality of the results 
when sound is recorded on a separate film are of course higher than 
when sound and picture are recorded together. The use of a light 

April, 1934] 



modulator with special immersional objectives permits great sim- 
plification of the entire installation and a record whose quality is the 
equal of that obtained with the heavy stationary systems. 

The original model of this type of equipment (Fig. 27) differed from 
that described in so far as the camera was driven inelastically through 
friction rolls and the apparatus mounted on a tripod even while record- 
ing on two separate films, to eliminate the necessity of interlocked 
motors and thus make possible the use of the camera motor exclusively. 
This model has not been generally used because of its inconvenience 
and certain defects of construction. 

FIG. 30. Schematic arrangement of Figs. 28 and 29. 

The stationary model of a portable set shown in Figs. 28, 29, and 
30 is one of our most recent designs of sound recording equipment, in 
which all the parts are compactly mounted on a common base- 
plate. Recording is made on the film, not while it is passing over the 
sprocket, but while it is passing freely between two cylinders. Such 
an arrangement greatly facilitates focusing the image on the film and 
observing the amplitude of vibration of the ribbon, which can be 
watched through the film by means of an observation microscope 
mounted in the top of the camera. To enable still more convenient 
control of the amplitude, a system has been used in which the form of 
a section of the sound wave is made continuously visible by means of 



[J. S. M. P. E. 

total internal reflections in prisms rotating immediately behind 
the film. Thus, not only the amplitude trend, but also an oscillo- 
gram of the sound can be seen in the microscope. 

How the film is threaded through the machine can be seen in 
Fig. 30 where A\ and A 2 are pull-down and hold-back sprockets, 
B is the cylinder stretching the film by friction between two cylinders, 
C and D, and directing the film to the point at which the sound is 
recorded, and E is the sprocket that drives the film, mounted on the 
shaft of the flywheel of the mechanical filter. The light modulator 
F is mounted on the base K of the camera: The modulator con- 
tains a rotating prism g of total internal reflection, permitting the 

FIG. 31. (Left) Newsreel recorder. 
FIG. 32. (Right) Schematic arrangement of camera of Fig. 31. 

rays of light to be excluded from the recorder and the modulator to 
be refocused without moving it from its place. The mechanical filter 
contains a cylindrical spiral spring and is air-damped; its natural 
frequency is only a quarter cycle per second. Every possible shock 
and vibration from the synchronous motor is absorbed by this filter. 
The ready removal of the upper case and the base-plate permits the 
apparatus to be compactly and conveniently packed in a trunk. 
All instruments for measurement and control are mounted on the 

Fig. 31 shows the model of a newsreel recorder in which the film 
driving mechanism, in order to reduce weight, is directly connected 

April, 1934] 



with the picture camera, and the mechanical filter, consisting of a 
spiral spring and a flywheel, rotates on the vertical shaft ^4 (Fig. 32). 
Fig. 32 shows the camera. Sound recording is done by a flashing 
lamp, in the housing C outside the film-drive housing, which il- 
luminates the drum B (Fig. 31) mounted on the shaft of the flywheel 
K of the filter. For use in focusing the image and installing the lamp 
an observing microscope is provided, with two oculars, one on each 
side of the camera. The objective is focused by the screw R. 
When sound is recorded for reproduction without a picture on the 
radio or in the theater, a large portion of the film is not used, and both 
this film and the chemicals used to develop it are wasted. Various 
methods have been developed for avoiding this waste by condensing 
the record on the film, which can be accomplished either by manifold 





D, ! 


a" - 


a. i j 

: : a 

a ' ; 

1 , a 



D- ' j j 

i : i a 

*^i j 

FIG. 33. Multiple-track methods 
for conserving film and chemicals. 

recording along the film or across it (Fig. 33). Both methods have 
been used with rather good results. 

One method of multiple recording along the film is to record on one 
edge of the film while it is moving in one direction, then reverse the 
camera and record on the adjacent strip, etc. Another method is to 
make a record in the form of a spiral on a closed loop of film. The 
first method can be accomplished very simply with an ordinary 
camera by reversing the light modulator and the direction of the film, 
but this does not produce an uninterrupted record. 

The second method, producing an uninterrupted record, can be 
accomplished by using a box permitting continuous movement of the 
film, as illustrated in Fig. 34. Through the flywheel the motor drives 
the cylinders of the box which rotate the ring of film. Fig. 35 shows 
schematically the arrangement of such a box, in which a radial system 
of cylinders and rollers is driven through conical gears 2 by the central 
conical gear 3. 



tr. a M. p. E. 

In the center of the box are placed sprocket 4 (see also Fig. 34) 
and similar sprockets at the edge of the box, both geared to the 
rollers. The diameters and angular velocities of rollers 1 and 
sprockets are identical. The rollers serve both to support and rotate 
the film roll. A crown resting on the rollers fixes the inner diameter of 
the film roll. The device is shown in Figs. 36 and 37. A special 
electro-frictional arrangement (2 in Fig. 36) shifts the drum that 
guides the film past the recording point just enough to prevent the 
successive sound tracks from overlapping. Metallic connections at 

FIG. 34. (Left) Scheme for producing an uninterrupted multiple-track re- 
cording, with continuous movement of the film. 

FIG. 35. (Right) Arrangement of the mechanism embodying the scheme 
of Fig. 34. 

the beginning and the end of the film loop control the action of the 
device, and disconnect the entire apparatus when all the space on the 
film has been used. The manufactured set can accommodate loops 
of from 2 to 1000 meters of film, and since eight tracks can be placed 
side by side, as much as six hours of uninterrupted sound can be 

In the first model that was designed, it was intended to provide a 
similar box for uninterrupted reproduction. Furthermore, it was 
expected that an optical system for sound reproduction could be 


included in the recorder. Experience showed that such a system was 
too clumsy, and accordingly a special arrangement (Fig. 38) was de- 
signed for reproducing from a manifold record. We have used this 
unit as an electro-stenographer for recording meetings, operas, con- 
certs, etc. Any portion of a record made by such a unit can be copied 
in a form suitable for insertion into any picture. 

The solution of the problem of manifold recording across the film 
would open tremendous possibilities in sound pictures, radio, teleg- 
raphy, telephony, and stenography, but the problem presents great 
difficulties. A special recording and reproducing arrangement is 
necessary. We work with systems of the type diagrammed in 
Fig. 39. 

FIG. 36. Photograph of multiple-track 
recorder of Figs. 34 and 35. 

Here the optical system produces on the film the image of a rec- 
tangular slit in the form of a thin line 2 to 2.3 mm. long and 0.02 to 
0.03 mm. wide parallel to the edge of the film. Ribbon N is also 
imaged on the film, covering half the length of the line when at rest. 
The rays forming the images reach the film by reflection from one or 
another of the 42 mirrors on the rotating drum L 4 . When L 4 is 
given Y 4 2 of a complete revolution, the rays reflected to it from one 
of the mirrors will move across the film from one edge to another 
(Fig. 40). On further rotation of the drum the rays will be re- 
flected from the next mirror and will sweep through the same path. 
By properly adjusting the speeds of the film and the drum the succes- 
sive sweeps of the rays will trace adjacent parallel paths across the 

228 A. F. CHORINE [J. S. M. P. E. 

film. Thus the mirror drum must rotate with a constant angular 
velocity that bears a fixed relation to the angular velocity of all 
sprockets driving the film. 

Fig. 41 shows the exterior of the apparatus for such recording. 
After being withdrawn from box 1 by the sprocket 2 the film forms a 
loop, and then is passed by the drum through frame 4 where the record 
is made. After forming another loop, it passes over drum 5 against 
which it is held frictionally by the bobbin, and finally is wound on 
the reel 8. To assure that the rotation of the mirror drum will be as 

FIG. 37. Another view of Fig. 36. 

uniform as possible, it is made of solid brass and carries a heavy fly- 
wheel on each side. 

The record can be conveniently reproduced by the same apparatus 
when the light modulator is replaced by a suitable optical system such 
as that shown in Fig. 42. This system projects on the film an image 
of the rectangular slit P in the form of a thin line 0.02 to 0.03 mm. 
wide and 2 to 2.1 mm. long. The light transmitted through the film 
Centers the photoelectric cell. 

In the operation of the system it may happen that before one mirror 



has completed reading its line, the next mirror begins to read the 
following line. To avoid this the mirrors in both recording and 
reproducing equipment must be very accurately mounted on the 
drum. Experience has shown that an error of L /4 angular minute in 
mounting a mirror will produce on the film an error of 0.012 mm. in 
the position where two consecutive lines begin. 

As the time for recording one line is VM second, the passage 
from one line to the other can introduce an unwanted signal with a 
fundamental frequency of 20 cycles per second. The usual amplifier 
does not pass such a frequency, but it will pass all the harmonics in- 

FIG. 38. Special arrangement of multiple-track recorder, used as 
an electro-stenographer for recording meetings, etc. 

eluding the second. The signal is of such form (Fig. 43) that these 
harmonics may be of considerable magnitude. The amplitudes of 
the first four harmonics, as the starting point of a line Si deviates 
from its proper position 5, have been calculated (Fig. 44) and show 
that it is difficult to avoid the production of harmonics. In general, 
the mirrors are mounted with an accuracy of 0.25 angular minute. 

Another remedy is used to eliminate the undesired frequencies in 
the reproducer. An additional lens L 5 (Fig. 42) is placed between the 
film and the photoelectric cell. Its focal distance is such that the 
distance between it and the film is sufficiently large to permit the 
rays that have passed through the film to diverge (Fig. 45). Near 


[J. S. M. P. E. 

FIG. 39. (Left) Experimental system for multiple recording across the film. 
FIG. 40. (Right) Illustrating the disposition of the sound tracks on the film. 

the lens an adjustable rectangular slit is placed. When the strip of 
light moves along the film from one edge to the other, the slit is slid 
in front of the lens L&. When the strips of light from two successive 
mirrors reach the slit, it cuts off portions of both strips in such a way 
that the light entering the photoelectric cell remains constant, 

By these two means the accurate mounting of mirrors and the 
diaphragming of the rays undesired noise due to the transition from 
one line to the next is considerably reduced. The largest volume of 
unwanted frequencies occurs during the silent portions of the record. 
Here the application of noiseless recording effects a further reduction. 
From Fig. 43 it can be seen that Ci/C equals 2, while with noiseless 
recording C\/C is reduced to 5/3; the amplitude of the harmonics is 

FIG. 41. 

Exterior view of apparatus for multiple recording across 
the film. 

April, 1934] 



reduced by an even greater amount. Further development of this 
multiple recording system is being directed toward achieving an 
arrangement giving greater light intensity. The automatic develop- 
ing and drying of cross-recorded film make the system extremely 
handy. Its only disadvantage is the high precision required in manu- 
facture. If further development can solve this difficulty, the system 
should find wide practical use. 

In order to produce a good sound record, the motion of the film 
during recording must be smooth. Our experiments to this end 

FIG. 42. (Lower) Arrangement for repro- 
ducing the multiple-track records across the 
film, recorded by the system of Figs. 39 to 

FIG. 43. (Upper left) Form of signal wave 
produced when passing from one line of a 
transverse recording to another. 

differed somewhat from the methods followed in other laboratories. 
A familiar cause of distortion is the inaccurate cutting of the teeth 
in gears, worms, etc. We used the apparatus illustrated in Fig. 46 
to measure inaccuracies in a gear train containing any number of 
gears from two up to those of an entire driving mechanism. 

The driving gear A is placed on a shaft whose turning angle can be 
measured with an accuracy of 4 angular seconds. On the shaft of the 
final gear B of the train is the mirror Ci from which a beam from the 
light source O is reflected to the rigidly fixed mirror Cz and thence to 



[J. S. M. P. E. 

the scale P. Turning the driving gear A turns the gear B, and shifts 
the position of the ray on the scale. After each turn, the mirror C\ 
is returned to its original position by an electromagnetic arrange- 
ment, so that it needs not be touched with the hands, and thus the 
ray on the scale returns approximately to its original position. 

On conducting a test, the starting point of the ray is first noted on 
the scale. The driving gear is then turned through the chosen angle 

FIG. 44. (Upper left) Amplitudes of the first four harmonics as the start- 
ing point of one transverse line of recording deviates from its proper posi- 

FIG. 45. (Right) Path of rays through lens L 6 of Fig. 42. 
FIG. 46. (Lower left) Method of checking inaccuracies in gear trains. 

and the new position of the ray noted on the scale. The mirror is 
now returned to its original position and the new zero point noted. 
Again the driving gear is turned through the same angle, and again 
the position of the ray is read. These operations are repeated to 
cover a complete cycle of the gear train. The angle through which B 
is turned can be read from the scale with an accuracy of 5 to 6 angular 
seconds. From the data the average angle through which B was 
turned, and the deviation of each turn from the average, can be cal- 

April, 1934] 



culated. Hence the percentage deviations and cumulative error, 
plotted in Figs. 47 and 48, respectively, can be obtained for a pair of 
ordinary cylindrical gears, and in Figs. 49 and 50 for cylindrical gears 
with oblique teeth. 

It is interesting to notice from the curves that the gears with 
the oblique teeth are the worst. This is easily explained by the con- 
ditions of their manufacture. The simpler the part, the more easily 
it can be made. Even the most inefficient worker could make the 

II tl'ia t2;!3 13:14 (415 18^16 H5.'rt 1718 18:19 192020 

ai'as aa^s zss* 2;as zslze aeis? 27 : ae 2912$ 29 so so 

FIG. 47. (Upper) Graph of percentage 
deviations of a pair of ordinary cylindrical 

FIG. 48. (Lower) Graph of cumulative 
error in a pair of ordinary cylindrical gears. 

ordinary gear sufficiently well, while the more complicated type of 
gear could be well made only by a highly trained mechanic. We 
have tested by this method all commonly used gear trains, and the 
results have always substantiated this practical rule. 

Another method used for measuring the uniformity of movement 
of such mechanisms employed a regulating microphotometer, and was 
used for testing cameras already assembled. Uneven motion of the 
recording sprocket (in early models a sprocket with 16 teeth was 
generally used) was caused by incorrectly cut teeth and incorrect 



[J. S. M. P. E. 

profile and marking of the sprocket driving the film. In these tests 
the film was wound around the driving sprocket and fixed with 
rubber bands, while the image of the slit, 0.02 X 3.0 mm., was fo- 
cused on the film. The sprocket was rotated at normal speed, and a 
special diaphragm exposed the sprocket to the image during exactly 
one revolution. By repeating the experiment with the film in the 
same position on the sprocket we could find out the periodic defect 

FIG. 49. (Upper) Graph of percentage 
deviation of a pair of cylindrical gears with 
oblique teeth. 

FIG. 50. (Lower) Graph of cumulative 
error of a pair of cylindrical gears with oblique 

in the cut of the drum, and by passing the developed film in front of 
the illuminated opening in the camera, we could visually analyze the 
movement of the mechanism. By ascertaining the gamma of de- 
velopment, and measuring the developed densities by Koch's 
microphotometer, we were able to obtain a quantitive curve of varia- 
tions in the drum speed. 

Figs. 51 and 52 are two typical records of the deflection of the rib- 
bon in the electrometer of the measuring device when the track de- 

April, 1934] 



veloped after such a test is microphotometered. Along the horizontal 
axis of the record 10 mm. is equivalent to 0.5 mm. on the film, or 
0.001 second. These particular records were made to test one of the 
first cameras manufactured for sound recording, which had just left 
the studio. It is apparent that inexact tooth shapes and poorly 
adjusted details have produced many small speed deviations. 


One of the problems in providing sound reproducing equipment for 
motion picture theaters was the design of a sound attachment that 
could be easily installed on the existing and most generally used silent 
picture projector of the "Tomp No. 4" type. The construction 

FIG. 51 . (Upper) Typical record obtained with registering 
microphotometer for determining variations of rotation. 

FIG. 52. (Lower) Similar to Fig. 51. 

adopted is shown in Figs. 53 and 54, where D is the housing of the 
sound attachment, which can be readily mounted under the table of 
the picture projector. The lower reel is attached to the lower wall 
of the housing, and inside is the mechanical filter consisting of a fly- 
wheel, flat spiral spring, and damping friction. The tube S, on the 
face, contains the optical system diagrammed in Fig. 55. To correct 
for the curvature of the image field produced by the objective, the slit 
G (0.23 X 21 mm.) has a curvature in the line perpendicular to the 
figure. A polished flat-parallel glass plate N, rotatable to ninety de- 
grees, and an observing microscope assist in focusing the image of 
the slit on the ribbon, and in checking the installation of the lamp 
wiit the aid of the opal glass C. The photoelectric cell and the pre- 

236 A. F. CHORINE [j. s. M. P. E. 

liminary amplifier tube are housed in a separate shielded container. 

The sprocket Y (Fig. 54) loops the film to prevent the transmission 
of shocks to the sprocket T from the lower reel. The reverse side of 
the housing has a removable cover on which is mounted a gear mesh- 
ing with gear X on the vertical shaft which thus connects the sound 
attachment with the top of the picture projector and drives the lower 
reel through a belt. The whole projector is driven by a single-phase 
synchronous motor, or by a d-c. motor with an electrical governor. 

This sound attachment was placed in use early. Many efforts 

FIG. 53. Sound attachment for existing 
silent projectors. 

have since been made to simplify the system, in order to reduce the 
cost in mass production, by eliminating the filter and simplifying the 
optical scheme, but so far without success. 

When sound films are reproduced in motion picture theaters, a gain 
control is often operated by some one in the auditorium in order to 
achieve a greater range of volume than can be reproduced directly 
from the film. Such manual control has, however, a number of de- 
fects : it depends upon the musical training of the man handling it, 
his attentiveness, and his familiarity with the picture; and it is diffi- 


cult to adjust the control for sounds of short duration, such as gun 
shots and other sudden noises. It is much more satisfactory to con- 
trol the gain automatically so that, when certain portions of the film 
pass through the apparatus, predetermined sections of a sound po- 
tentiometer are connected to the amplifier. 

To attain such control a special punch is used to press into the film 
spherical embossings about 3 mm. in diameter and 0.75 to 1 mm. 
high. As many as five such embossings can be placed across the film, 
between frames so that they can not be seen in the picture projected 
on the screen. Mounted on the projector, somewhat above the pro- 

* T 

FIG. 54. Close-up of sound projector of Fig. 53. 

jection window, are five small levers 0.2 mm. from the film. When 
the film passes through the projector (Fig. 56) an embossing moves 
its corresponding lever K, and so through a pair of spring contacts M 
and a relay in box L connects the corresponding section of the poten- 
tiometer to the amplifier. This section remains connected until an- 
other embossing acts. By mounting the levers so that the emboss- 
ings act while the frame is stopped during projection, the duration 
of the current pulses operating the relays is made 0.035 to 0.037 sec- 
ond, which is quite sufficient for reliable relay work. 

In the first model of the volume control unit, four groups of con- 
tacts were used, providing complete silence by switching off the 



[J. S. M. P. E. 

sound, and three values of sound intensity. The fifth contact group 
was added to switch the sound from one group of loud speakers near 
the screen to another group located elsewhere in the auditorium. 
Thus, for instance, a conversation can be reproduced between an ac- 
tor on the screen and a supposed partner somewhere in the auditorium 
and the sound of a falling object apparently thrown out from the 
screen can be made to emanate from the auditorium. With the same 
five pairs of contacts and a more complicated relay system, some 
thirty different operations can be performed by various combinations 
of operated and unoperated contacts. In such a way the lights in 
the auditorium can be turned off, operation can be switched to an- 
other projector, etc. 

FIG. 55. Optical system contained in tube S of Fig. 54. 


The work already described, on optico-mechanical apparatus for 
recording and reproducing sound, has been accompanied by the de- 
velopment of amplifying equipment, beginning with simple battery- 
operated amplifiers similar to those used in radio, and ending with 
equipment fulfilling the requirements specifically demanded by the 
sound picture. Since the recording of sound and its transmission to 
the amplifying equipment are subject to various requirements that 
can not be satisfied by one universal equipment, separate units have 
been designed for recording and for reproduction. 

Amplifying equipment for sound recording on film consists of the 
following parts: mixer equipment for three microphones with pre- 

April, 1934] 



liminary amplifier; final amplifier for the recording; equipment for 
diminishing the noise of the film, and the monitoring amplifier. 

The mixer equipment is designed to accommodate three micro- 
phones of the condenser or moving-coil type. It includes three po- 
tentiometers, for varying the volume of each microphone through a 
24-db. range in steps of 1.5 db.; a three-stage preliminary amplifier 
with a gain of 40 db.; a regulator for varying the output volume 
through a range of 32 db. in steps of 2 db.; and power control instru- 
ments and a volume indicator. This equipment is fed from the same 
battery that feeds the microphone amplifiers, and is screened to re- 
duce interference from outside power sources. The outputs of the 

FIG. 56. Automatic volume control system, operating 
by means of embossings on the film which actuate spring 
contacts and relays. 

microphones are connected to the potentiometers without the use 
of coupling transformers to guard still further against the interference 
that might be picked up by a non-toroidal retard coil or transformer. 
The mixer equipment is placed close to the scene being photographed, 
so that the operator may watch the scene and at the same time moni- 
tor the sound through a telephone. 

Other parts of the recording amplifier are placed on a standard 
rack. They are fed entirely by alternating current, through recti- 
fiers mounted on the same rack. Normally all this is placed in a sepa- 
rate room. When necessary, each part can be removed from the 
rack, packed in a separate box, and used on location either separately 
or in combination with any other part. 

240 A. F. CHORINE [j. s. M. P. E. 

The complete recording amplifier not only increases the volume to 
a degree sufficient to operate the light modulator, but also, by a spe- 
cial network, corrects the distortions of frequency characteristics that 
result from the finite width of the slits in the recorder and reproducer, 
and from the limited resolving power of the film. The former can be 
calculated; the latter varies somewhat with the chemical treatment 
of the film. Therefore, when the corrective network was designed 
it was given a frequency characteristic that would compensate for the 
effect of the slits and the average effect of the film; this characteristic 
has been maintained with an exactitude of 1.5 db. 

The equipment for reducing the film noise consists of a two-stage 
amplifier, a copper-oxide rectifier, and a filter that passes only the 
d-c. component brought to the ribbon of the recording modulator and 
which changes the position of the ribbon in accordance with the am- 
plitude of the signal. 

The complete amplifier and the silencer are fed from a common 
rectifier, because normally both are in operation at the same time. 
When necessary, as, for instance, on location, batteries can be used. 
The monitoring amplifier amplifies a portion of the signal output of 
the complete amplifier, enabling the recorded sound to be heard as it 
should be when reproduced. This amplifier is also operated on alter- 
nating current, and has its own rectifier. 

All the amplifiers described, except the distortion-correcting stage, 
have frequency characteristics that are flat within 1 db. between 50 
and 10,000 cycles per second. Distortion is corrected at frequencies 
up to 7000 cycles per second, above which the distortion varies too 
greatly with the way the film is manufactured to make correction 

Since the correction of distortion originating anywhere between 
the microphone and the loud speaker is accomplished in the recording 
amplifier, the reproducing amplifier is given a frequency characteris- 
tic that is flat within 1.5 db. between 50 and 10,000 cycles per sec- 
ond. All operating potentials, including that for the photoelectric 
cell, are obtained from 50-cycle a-c. sources. To reduce the noise 
from the sources the rectified anode potential is carefully filtered, and 
compensations for the noise are introduced. The noise is thus re- 
duced to less than 0.1 per cent, and actually is imperceptible. 

The equipment consists of four parts : a photoelectric cell placed on 
the picture projector; a preliminary single-stage amplifier with a gain 
of 27 db., placed beside the picture projector on the wall of the camera ; 


an amplifier with a 79-db. gain and maximum undistorted power out- 
put of 2.5 watts; and an amplifier with a 20-db. gain and maximum 
undistorted power output of 30 watts. In small theaters the fourth 
element is omitted and the total gain is restricted to 106 db. The 
photoelectric cell and the preliminary amplifier are fed from a 2.5- 
watt amplifier. When the 30-watt amplifier is used, the 2. 5- watt 
amplifier works into the power amplifier and the control loud speaker, 
and the total gain is of the order of 126 db. 

To simplify operation, the amplifier is transferred from one pro- 
jector to another by means of relays operated by pressing buttons 
placed near each projector and on the amplifier itself. Pressing a 
button also lights pilot lamps indicating the projector with which the 
amplifier is connected. Since in vacuum tubes of the heater type the 
plate potential must not be applied until the cathodes have been suf- 
ficiently heated, the power supply circuit includes a relay that auto- 
matically delays the application of the plate potential for the neces- 
sary interval of 40 to 60 seconds. 

On the remainder of the apparatus used in sound recording and 
reproduction, we have done little development work. We have used 
many types of microphones, both of our own and foreign construction 
and have found their differences perceptible but not of conclusive im- 
portance. For loud speakers we have used the normal electro- 
dynamic type. Horn loud speakers have not yet been distributed in 
great numbers, although we have at our disposal both our own and 
certain foreign types. In practical use we have generally employed 
photoelectric cells of the potassium type, gas-filled, but we have ex- 
perimented with various types of caesium and rubidium elements both 
gas-filled and vacuum. 

In concluding this brief review, the author wishes to express his ap- 
preciation and gratitude to his close associates for their enthusiasm 
and assistance, past and present. Professors Litvinsky, Smerinin, 
Kulikoff, Yakhontoff, and Nikolsky, and Engineer Podkovsky have 
been associated with the work from the outset; Engineers Borissoff, 
Vorobioff, Lessnikoff, Moshonkin, Obukhoff, Polansky, Stepanoff, 
Salier, Timartzeff, Ussikoff, and Chibissoff joined it later; and Engi- 
neers Volkoff, Mukhatchoff, Shtzo, Kumitz, and V. V. Petroff have 
been of great help in the practical work in the studios, theaters, radio 
centers and the workshops of Soyuzkino. 


S. K. WOLF** 

Summary. The extension of the frequency and volume ranges in recording and 
reproducing sound has aroused a greater and more critical consciousness of the 
importance of theater acoustics. It follows that higher fidelity in reproduction excites 
greater intolerance of the needless distortion caused by poor acoustics of the theater. 
To cope with the new 'situation, engineers have developed new instruments for acoustical 
analysis, which provide greater precision and facility in detecting defects and in 
determining the necessary corrections. 

In addition to instrumental developments there have been concurrent advances in 
acoustical theory and practice. The result is that the more stringent requirements 
imposed on the acoustics of the theater by the enlarged frequency and volume ranges 
can be fulfilled adequately and practically. The paper discusses the requirements 
and describes some of the available methods for complying with them. 

The extension of the frequency and volume ranges in recording 1 and 
reproducing 2 sound has brought about a greater and more critical 
consciousness of the importance of quality as a factor in sound 
pictures. It may well be said that, as one result, sound pictures have 
come into their majority and achieved their birthright; they are now 
a medium of entertainment free from the necessity of leaning on their 
novelty as an apology for their deficiencies. 

It follows naturally that greater fidelity of reproduction excites 
greater intolerance of the needless distortion of quality caused by 
poor acoustics in the theater. With every other link in the chain 
from the recording set to the theater made as nearly perfect as modern 
engineering can make it, it would appear a perverse blow indeed if, 
through either negligence or ignorance, the character of the repro- 
duction were degraded by improper acoustical conditions. 

The extension of the frequency and volume ranges has, of course, 
imposed additional requirements on the acoustics of the theater; but 
fortunately concurrent advances have been accomplished in acous- 
tical theory and practice, as well as new developments in instruments 

* Presented at the Fall, 1933, Meeting at Chicago, 111. 
** Electrical Research Products, Inc., New York, N. Y. 


for making analyses. The result is that both the old and the new re- 
quirements can be met adequately and practically and at the same 
time with greater precision and speed than in the past. This paper 
discusses the new situation and describes some of the available meth- 
ods of coping with it. 


The work done by Sivian, Dunn, and White 3 and by W. B. Snow 4 on 
the spectra of musical instruments, and that done by Fletcher on 
speech spectra, indicate the importance of the higher and lower fre- 
quencies in the reproduction of music and speech. It follows that, if 
these additions to the frequency bands are worth reproducing, they 
are worth reproducing correctly. It is, therefore, necessary to in- 
clude an analysis of the acoustics of the theater relative to those added 
frequencies when making acoustical corrections. 

One of the most important factors in auditorium acoustics is rever- 
beration. As is generally known, reverberation is determined by the 
acoustical absorption present in an auditorium and the cubical con- 
tent of the auditorium. Through the work of W. C. Sabine, 5 Lif- 
schitz, 6 and MacNair, 7 we know that there are optimal conditions for 
reverberation which vary with frequency. Reverberation times in 
excess of the optimal lead to excessive "liveness," as the sounds are 
prolonged beyond the proper length of time. Reverberation times 
less than the optimal lead to a characteristically "dead" and "flat" 
quality. It has been found that there is a median band of reverbera- 
tion times within which good conditions obtain. 

In Fig. 1 are shown optimal reverberation times as a function of 
the frequency, for an auditorium of 300,000 cubic feet. The cross- 
hatched sections at the two ends represent the extensions involved in 
wide-range reproduction. It will be seen that more reverberation 
is acceptable at low frequencies, due to the fact that the human ear 
is less sensitive to low-frequency sounds. Theaters that were acousti- 
cally satisfactory prior to wide-range reproduction may require ad- 
ditional treatment in order to avoid the confusing and hollow quality 
deriving from excessive low-frequency reverberation. 

Acoustical correction at such low frequencies is complicated by the 
fact that in most cases the absorption of acoustical materials has not 
been tested at frequencies below 128 cycles, due to the lack of ade- 
quate testing facilities in the past. It is not always possible, there- 
fore, to determine the conditions existing in a theater merely by com- 



[J. S. M. P. E. 

puting the reverberation times, as has been the general practice. It 
is safer to make actual measurements in the theater to ascertain the 
true picture and thus arrive at the proper recommendations for cor- 

Fig. 1 indicates that frequencies above 4000 cycles should also be 
included in making a thorough acoustical analysis. We know, from 
an analysis of vocal and musical spectra, that such frequencies are of 
great importance in imparting brilliance and character to the repro- 
duction. Here, also, correction of a theater is complicated by the 
fact that absorbing materials have usually not been tested above 4096 

too tooo 


FIG. 1. Optimum reverberation times for 300,000 cu. ft. theater. 

cycles; and, what is worse, that at the present time they are not being 
tested above 2048 cycles by most manufacturers. Another compli- 
cation is introduced by the absorption of the air as the sounds travel 
through it. Air absorption is very appreciable at high frequencies, 
and becomes more so as the frequency is increased. It depends upon 
various conditions such as temperature and humidity, which constitute 
a source of uncertainty (excepting in air-conditioned rooms) in mak- 
ing corrections. However, experience has enabled us to make thor- 
oughly reasonable allowances for the possible variation of atmos- 
pheric conditions. 

The reverberation times shown for the high frequencies in Fig. 1 
were determined on the basis of MacNair's work. Their attainment 
in the theater on the basis of computation alone is uncertain; and, 
again, for a higher degree of accuracy direct measurements must be 
made. However, in a great many instances, measurement is im- 



practicable for commercial reasons and, therefore, compromise solu- 
tions must be adopted. 

In Fig. 2 are shown reverberation times measured in two auditori- 
ums, indicating the extremes that may exist. One is far too "live" 
at high frequencies, the other at low frequencies, the over-all char- 
acteristics indicating intolerably bad situations. It is true that they 
represent extreme cases; but, at the same time, they indicate what 
may happen when no precautions are taken to achieve suitable and 
proper acoustical conditions. 

The extension of the low-frequency range has accentuated another 
factor which occasionally gave trouble in the past : resonance. Room 
resonance, usually at comparatively low frequencies, causes a pro- 





FIG. 2. Reverberation frequency characteristics: two theaters of 300,000 
cu. ft., compared with optimum time band. 

nounced "boomy" quality, with disagreeable emphasis and prolonga- 
tion of certain tones beyond their normal values. A loud speaker 
that has a comparatively good frequency response characteristic when 
measured in a "dead" room may exhibit a humped and jagged char- 
acteristic when installed in a theater. Resonance may exist between 
the rear wall of the stage and the speaker baffle if their planes are 
nearly parallel. In the present era of large baffles, such a situation 
can be serious, and in a considerable number of instances has re- 
quired correction. However, the condition can easily be remedied 
by installing absorbing materials on the surfaces that cause the 
trouble, Sometimes satisfactory results can be attained by re-aligning 
the loud speaker baffle. Cases of mechanical resonance would require 

246 S. K. WOLF [j. s. M. P. E. 

special consideration and treatment. Tests of one installation showed 
resonance at 80 cycles. By treating the surfaces surrounding the loud 
speaker with 2-inch rock wool, the sound intensity behind the baffle 
was reduced by as much as 8 decibels at a point adjacent to the baffle 
and 23 inches from the speaker unit. 

Such resonance should not be confused with excessive low-frequency 
reverberation, which is similar in some respects. Resonance causes 
an accentuation of a few discrete frequencies, and is the more trouble- 
some owing to the fact that any sort of impulsive tone will lead to the 
appearance of those frequencies Accordingly, the reproduction of 
speech or music may be marred by the intermittent accompaniment of 
such resonances. Low-frequency reverberation, on the other hand, 
affects all frequencies within the band, and will appear as a rolling 
prolongation of the low-pitched sounds actually present in the passage 
being reproduced. Its correction involves the installation of absorb- 
ing material in the proper amounts on surfaces throughout the theater 
and not total suppression as in the case of resonance. 

Because of the fact that multiple sets of loud speakers are required 
to embrace the extended frequency range, the problem of achieving a 
uniform distribution of acoustical power at all frequencies may be re- 
garded as somewhat simplified. The loud speaker units and their 
associated circuits at present divide the frequency range among them- 
selves into either two or three bands. Since each unit serves a rela- 
tively narrow band width, the directive properties of the combination 
are more nearly uniform throughout the range, and thus better dis- 
tribution is achieved. However, because of the over-all improved 
quality, smaller irregularities become more prominent, so that even 
with a more amenable system, a great deal of care is still necessary. 
Further discussion of this phase of the acoustic problem is beyond the 
scope of this paper. 


Extension of the recorded and reproduced intensity range is not so 
recent a development but, nevertheless, it may fittingly be treated 
here. The recorded range has been increased by about 8 decibels to 
an amount more nearly capable of accommodating the intensity varia- 
tion encountered in the original production of speech and music. 
This increase, brought about by "noiseless recording," is accom- 
plished entirely at the low-intensity end of the range, the maximum 
values being untouched. Accordingly, it is now possible to hear 



relatively faint sounds that would formerly have been submerged in 
a welter of surface noise. In order to foster and protect this gain, it 
is necessary to guard against excessive ambient theater noises. 

The effect of noise 9 in reducing the benefits of wide volume range 
recording is shown in Figs. 3(a) and 3(6), which were obtained with 

TIM irratVAl 18 KUDUS 

FIG. 3. (a), Variation of reproduced sound level of a 
wide-range recording under quiet acoustic conditions; (&), 
same as (a) but reproduced in the presence of acoustic 
noise of 41.5 db. 

an automatic level recorder, developed by the Bell Telephone Labora- 
tories. Fig. 3 (a) is an intensity level chart of a recording having a 
wide volume range with no external noise present. It will be seen 
that all low-intensity sounds are clearly defined, and that the full 
volume range could be enjoyed by an auditor without requiring close 
aural attention. Fig. 3(b) is a chart of the same recording under ex- 

248 S. K. WOLF [J. s. M. P. E. 

actly the same conditions, with the same upper intensity level, but 
reproduced in the presence of noise. It is apparent that the effective- 
ness of the low-intensity range has been seriously impaired by the 
presence of the noise, with a resultant loss of intelligibility and enjoy- 
ment. Of course, even in the presence of the noise, the average audi- 
tor could still distinguish and interpret speech and music, but only 
with difficulty and with closer attention. For such reasons it is as 
much the duty of the exhibitor to provide good hearing conditions 
for the aural comfort of his patrons as it is to provide good ventilation 
and seating for their physical comfort. 

The maximum permissible limit of noise in the theater is deter- 
mined by the noise produced by the audience. It is possible to treat 
other noise sources but the audience is a source over which our con- 
trol is meager. From a series of observations in a number of theaters, 
it has been found that the noise level that may be considered to be 
representative of a comparatively quiet audience is approximately 
35-40 decibels above the threshold of audibility. It is to be appre- 
ciated that greater levels may occur momentarily, and that during 
tense dramatic moments they may be much lower. It is apparent, 
then, that the audience noise is the controlling factor for the maximum 
permissible noise level and that every effort should be made to reduce 
all theater noises to levels lower than it. That does not imply that 
the noise sources will not be heard, but rather that their effects will 
have been overcome from a practical standpoint. 

Theater noises may be divided into two general classes: those 
produced externally of the theater, and those within the theater and 
incidental to its operation. Primarily among the noises produced 
externally are those emanating from street traffic, industrial estab- 
lishments, and railway or other forms of transportation. Observa- 
tions conducted in New York City have shown that the average 
street noise level due to traffic is approximately 70-80 decibels, de- 
pending upon the nature and density of, and proximity to, the traffic. 
Momentary peaks, of varying duration, of 90-95 decibels may be 
encountered, and in a few isolated cases levels of street noise as high 
as 105 decibels have been measured. It is apparent, therefore, that 
in order to remain below the maximum internal noise level, the trans- 
mission-reduction factor of the theater structure should lie between 
35 and 45 decibels to overcome average traffic noise for satisfactory 
conditions; and, in very severe situations, between 60 and 70 decibels. 
In suburban locations the average street noise level may be consider- 



ably less than the values here presented, but the momentary peak 
values stated will probably be still representative. 


To supplement and augment their theoretical information, engi- 
neers have had to seek the aid of instruments. As has been the case 
with most work in acoustics, recent or otherwise, and because the 
field is a comparatively unexploited one, instruments have had to be 
developed especially for the purpose. One of the devices that has 
just become available is the high-speed level recorder, 10 which has 

FIG. 4. High-speed level recorder. 

already proved its value in conducting measurements of reverbera- 
tion time and frequency response. Its name suggests its purpose: 
it automatically records sound intensity levels as they fluctuate at a 
point. The curves in Fig. 3 were obtained by means of one type of 
such a level recorder. 

Fig. 4 is a photograph of the meter, designed by the Bell Telephone 
Laboratories, set up for synchronous operation with a beat-frequency 
oscillator. Excluding the oscillator, there are three separate units: 
the recording unit proper, its associated amplifier, and a battery box. 
The record is impressed on a moving waxed paper strip by a stylus 
which follows the changes of the sound intensity. The speed of the 
paper may be varied in three steps from 3 /c4 inch per second to 3 
inches per second. The stylus also may be adjusted to follow changes 



[J. S. M. p. E. 

of intensity from 45 decibels per second to as much as 360 decibels per 
second. By driving the paper and the oscillator frequency control 
synchronously, the horizontal axis may be made proportional to the 
frequency instead of to the time. 











FIG. 5. Acoustic response measuring equipment. 

Fig. 5 is a schematic drawing of the recorder as it is used for de- 
terminations of loud speaker response in auditoriums. It is obvious 
that by means of such an arrangement measurements can be made 
expeditiously and automatically and that, therefore, very complete 
information concerning the performance of a loud speaker in a given 
auditorium can readily be obtained. Fig. 6 is an example of such a 


FIG. 6. Typical frequency response characteristic. 

study. The sound intensity level at a point in a theater is given as 
a function of the frequency from to 10,000 cycles. It will be noted 
that the response falls off above 4000 cycles. In this particular in- 
stance, the loud speaker input circuits were adjusted to produce such 
an effect in order to avoid accentuation of the (high-frequency) sur- 
face noise very prominent in the older recordings. By conducting 



several such tests at representative points in a theater, valuable infor- 
mation concerning the distribution of sound energy, the over-all re- 
sponse of the system, resonances, and peculiarities of the auditorium 
may be ascertained. 

The level recorder is useful also for measuring the rate of growth 
and decay of the sound energy. At its most sensitive setting it can 
record reverberation times as short as 0.028 second, which is shorter 
than is usually attained even in "dead" rooms. In Fig. 7 are four 

FIG. 7. Sound energy growth and decay curves: (A) 160 cycles 
per second; (B) 500 cycles; (C) 1000 cycles; (D) 2250 cycles. 

examples of growth and decay curves. It will be observed that the 
sound intensity does not decay uniformly but, rather, rises and falls 
at about an average rate. In extreme cases this fact may be substan- 
tiated by direct aural tests, but usually instrumental observations 
are necessary to demonstrate the existence of such peculiarities. 

Because the scope of this paper is limited, only brief mention will 
be made of other instruments. The sound meter used by us for 
studying noise levels has been described before. 8 ' 9 It is of great value 

252 S. K. WOLF 

in arriving at recommendations for treating troublesome noise sources 
in theaters and on recording stages. Refinements in design have en- 
abled us to produce such a meter consisting of only one comparatively 
light case containing all the needed equipment. Another instrument 
has also become available : a precision analyzer, also designed by the 
Bell Telephone Laboratories. This device is useful in analyzing many 
different types of noise, testing the linearity of response of amplifiers 
and loud speakers, and supplementing other instruments when it is 
necessary to isolate given frequency bands. 


1 HOPPER, F. L,: "Wide-Range Recording," /. Soc. Mot. Pict. Eng., XXII 
(April, 1934), No. 4, p. 253. 

2 WARD, J. S., AND WILLIS, F. C. : "Wide-Range Sound Reproduction," 
presented at the Fall, 1933, Meeting of the Soc. Mot. Pict. Eng. 

3 SIVIAN, J. L., DUNN, H. K., AND WHITE, S. D.: "Absolute Amplitudes and 
Spectra of Certain Musical Instruments and Orchestras," /. Acoust. Soc. of 
Amer., II (Jan., 1931), No. 3, p. 330. 

4 SNOW, W. B.: "Audible Frequency Ranges of Music, Speech, and Noise," 
/. Acoust. Soc. of Amer., Ill (July, 1931), No. 1 (part 1), p. 155. 

5 SABINE, W. C.: "Collected Papers on Acoustics," Harvard Univ. Press, 
Cambridge, Mass., 1927. 

6 LIFSCHITZ, S. : "Mean Intensity of Sound in an Auditorium and Optimum 
Reverberation," Phys. Rev., 27 (May, 1926), No. 5, p. 618. 

7 MACNAIR, W. A.: "Optimum Reverberation Time for Auditoriums," 
J. Acoust. Soc. Amer., I (Jan., 1930), No. 2, p. 242. 

8 KNUDSEN, V. O. : "Measurement and Calculation of Sound Insulation," 
/. Acoust. Soc. Amer., II (July, 1930), No. 1, p. 129. 

9 WOLF, S. K., AND TWEEDALE, J. E.: "Theater Noise Problems," /. Soc. 
Mot. Pict. Eng., XIX (Dec., 1932), No. 6, p. 499. 

10 WENTE, E. C., BEDELL, E. H., AND SWARTZEL, K. D.: "A High-Speed 
Level Recorder for Acoustical Measurement," presented at the May, 1933, 
Meeting of the Acoust. Soc. of Amer. 


Summary. The recent improvements in sound quality resulting from the ex- 
tension of the frequency and intensity ranges are the results of coordinated activity 
in recording equipment and processes, reproducing equipment, and theater acoustics. 
This paper discusses the recording phase of the process. A wide-range recording 
channel consists essentially of the moving-coil microphone, suitable amplifiers, a new 
recording lens, and certain electrical networks. 

The characteristics of such a system, from the microphone to and including the 
processed film, are shown. Other factors fundamentally associated with wide-range 
recording, such as monitoring, film processing, the selection of takes in the review 
room, and re-recording, are also discussed. The changes brought about by this system 
of recording result, first, in a greater freedom of expression and action on the part of 
the actor; and, second, a much greater degree of naturalness and fidelity than has 
been previously achieved. 

Since the advent of sound in the motion picture industry, the 
sound engineer has steadily endeavored to make the reproduced sound 
more natural and pleasing. While there has been continuous effort to 
accomplish such improvements, the results have been apparent to the 
public only intermittently, as in the case of noiseless recording, and 
in the more recent improvement in sound quality known as "wide 
range." The latter step, which is basically an extension of the range 
of frequency, 1 both in recording and reproduction, has been achieved 
by coordinated activity in three fields : that of recording equipment 
and processes, that of reproducing equipment, and that of theater 
acoustics. The improvements in the latter two fields are described 
elsewhere, 2>3 and this discussion will therefore be confined to the 
changes made in the recording channel to attain this most desirable 

As the basis of coordinated design, it has been agreed that the over- 
all characteristic of film recording and reproducing equipment should 
be essentially uniform at all frequencies. However, due to condi- 
tions affecting the sound previously to pick-up, and other conditions 

* Presented at the Fall, 1933, Meeting at Chicago, 111. 
**^ Electrical Research Products, Inc., New York, N. Y. 




[J. S. M. P. E. 

affecting it subsequently to its leaving its reproducing apparatus, the 
characteristic may, at times, have to be slightly modified in order to 
achieve the most pleasing effect. 

The changes made in equipment and their result on the circuit and 
on the sound will be discussed in the order of their occurrence in the 
recording channel, beginning with the pick-up. The moving-coil 
microphone, which appears to be the most suitable pick-up instru- 
ment, has been described in detail elsewhere. 4 Its use results in a 
much truer reproduction of the original sound than was attained 
with other microphones used in the past. Practically, its use for dia- 
log has resulted in a greater freedom of dramatic action, making it pos- 
sible to produce takes that would have been unsatisfactory with mi- 




EQUAL/2 Eft 






60 fOO 1000 


FIG. 1. Recording channel characteristics. 


crophones of the older types. Probably a part of this additional 
freedom is due to the better high-frequency response of the moving- 
coil microphone, resulting in good articulation even though the actor 
move freely about the set. 

While the frequency response of the amplifier system is uniform 
over the recording range, modification of the characteristic is fre- 
quently necessary to achieve the most pleasing effect. These modi- 
fications are effected by a "dialog equalizer," which is introduced 
into the recording system during the process of recording the dialog. 
Its purpose is to attenuate the low-frequency speech currents gradu- 
ally so as to reduce the "chesty" and "panel resonance" effects that 
become distinctly so noticeable on extending the low-frequency 
range. The attenuation required is not definitely fixed, since it is 
frequently desirable to alter it to suit varying conditions on the set. 
Its purposes are: 

(1) To diminish excessive low-frequency reverberation or resonance, fre- 
quently encountered in 'sets. 

April, 1934] 





(2) To compensate for an apparent increase in the low-frequency response 
when the speech is reproduced at a greater volume than that obtaining during 
the recording. 

(3) To effect psychological compensation in scenes that would obviously be 
incompatible with their appearance in a large theater; as with an intimate close- 
up in a large auditorium, with the source of sound at some distance from the 
audience, in which case the desired feeling of intimacy can not be easily aroused. 

From this it is apparent that the dialog equalizer is not necessary 
for musical recordings, and may not be required for certain kinds of 
speech, such as those that are declamatory. 5 In addition to the 
dialog equalizer, a high-pass filter is sometimes required to exclude 
extraneous low-frequency sounds occurring on the set. 

The characteristic of the recording channel as affected by a dialog 
equalizer is shown in Fig. 1. This 
particular equalizer droops more at 
the low frequencies than most equal- 
izers now in use. On the same fig- 
ure is shown also the effect of the 
low-pass filter used to adjust the 
final characteristic for the upper fre- 
quencies. This is perhaps better 
shown in Fig. 2. 

The light-valve used with the 
system is tuned to 9500 cycles, and 
its r,ising characteristic in the re- 
gion immediately below resonance 
serves the useful purpose of off- 
setting the film loss and the loss 
due to the ribbon velocity effect. 6 
Heretofore the characteristics of 
light valves did not rise sufficiently 
to accomplish such a purpose, and a 
film compensating network has in 

some cases been used to balance the residual recording losses. 
However, the new valve characteristic, assisted by an improve- 
ment in the design of the recording lenses, is now sufficient to 
compensate for all recording losses, and the film compensating net- 
work becomes unnecessary. Returning to Fig. 2, the film char- 
acteristic shown is the one obtained with standard positive film, 
using the improved lens system. This figure shows also the com- 





1000 000 


FIG. 2. Light-value, film, and re- 
cor ding system characteristics. 



[J. S. M. P. E. 

posite curve representing the net effect on the light-valve character- 
istic, the film loss, and the characteristic of the recording system, in- 
cluding the low-pass filter. In order to obtain such a characteristic 
it is essential that the film be processed carefully and that no appre- 
ciable slippage shall occur in printing. With the development of new 
film emulsions, better resolving power will probably result, so that a 
film recorded with this valve might have a rising frequency charac- 
teristic. This will permit the light- valve to be tuned to a higher fre- 
quency, and thus allow a further extension of the frequency range; or 
it would permit using the present recording system with some attenua- 
tion of the high frequencies in reproduction, thus effecting a reduc- 
tion of the relative noise level. Combining the final characteristic 
with that of the recording channel, shown in Fig. 1, we have Fig. 3, 
which is the over-all characteristic of the wide-range recording system 



FIG. 3. Combination of characteristics of Figs. 1 and 2. 


as viewed by the photoelectric cell in the reproducing equipment. 
This characteristic includes the recording amplifiers, low-pass filter, 
light-valve, and all film losses. For purposes of comparison, a similar 
characteristic is shown of the recording system as previously used. 

However, that is not the complete story, as it neglects the effect 
of the microphone. Fig'. 4 shows the final result for both systems. 
The characteristic of the wide-range system includes that of the mov- 
ing-coil microphone, while that of the previous system includes that 
of the condenser microphone which was used with the older type of 
recording channel. It will be seen that the wide-range system is much 
more uniform in frequency response over a wider band of frequen- 
cies than the previous system. 

April, 1934] 



The new system may be said to have its upper cut-off at a frequency 
of about 8000 cycles. The low-frequency cut-off is not fixed, having 
been made, as already explained, purposely indeterminate in order 
that it may be changed to suit varying conditions. In music, for ex- 
ample, the cut-off is placed below the lowest frequency that the repro- 
ducing system is expected to reproduce, while for speech the lower 
frequency cut-off or droop is so placed that the most pleasing final 
effect is obtained. 

Under the former conditions of recording there existed a deficiency 
in frequency components above 5000 cycles. This lack of high fre- 
quencies was partially compensated by the accentuation of fre- 
quencies in the region of 3500 cycles by the condenser microphone. 
This accentuation resulted in a decided harshness and a nasal quality 
that are absent in wide-range recordings. 


60 100 


FIG. 4. Final characteristics including microphone. 


Further extension of the frequency range in recording will probably 
require higher tuning of the light- valve and an improvement in film 
characteristic. Valves having tuning points well above 12,000 cycles 
have already been used, and certain special test emulsions having the 
necessary extended response have been produced. The limitation of 
valve tuning follows from the presence of components whose fre- 
quency is such as to cause valve overload. The low-frequency re- 
sponse is limited by the loud speaker system and certain peculiarities 
of noise reduction. Improvements in the noise reduction system 
may, in the future, off-set this to some extent. Improvements in 
film emulsions, resulting in a greater signal-to-noise ratio are, of 
course, most desirable. 

The use of the moving-coil microphone has made it necessary 

258 F. L. HOPPER [j. s. M. P. E. 

to make further modifications in the noise reduction system, particu- 
larly for the higher frequencies. These changes do not affect the 
over-all frequency response of the system, but do add materially to 
its effectiveness of operation. 

Monitoring facilities for wide-range recording assume added im- 
portance, since the mixer must be provided with equipment enabling 
him to hear all the sounds being recorded that will subsequently be 
reproduced in the theater. The electrical characteristics required 
for the monitoring system are equivalent to those of the recording 
channel as to fidelity; and, in addition, improved loud speakers have 
been provided for monitoring. If the older monitoring facilities were 
to be used with a wide-range system, many sounds or extraneous 
noises might be recorded on the film that had not been heard by the 
mixer, thus making retakes necessary. Such improved facilities are 
particularly desirable when recording music, since judgment as to 
orchestral balance and instruments employed may be different when 
using a monitoring system whose frequency range is limited. 

That a satisfactory original record of the sound be made is not suf- 
ficient; three other steps remain: the selection of "takes" in the re- 
view room, the re-recording (when done), and the proper processing 
of the release print. The importance of the proper review room and 
associated equipment is even greater than the importance of ade- 
quate monitoring. The system should, of course, reproduce the en- 
tire range of frequency that has been recorded, and the acoustics must 
be so adjusted that the effect created in the review room is as nearly 
representative as possible of the effect that will be obtained in the 
better theaters. As it is impossible for a small review room to have 
the same acoustic properties as a theater, final judgment as to the re- 
produced sound must be formed by playing the film in a theater that 
is satisfactory for wide-range reproduction, as discussed in the papers 
previously mentioned. 

As in most studios the "dailies" are either partially or wholly re- 
recorded before the release prints are made, a re-recording channel 
must be provided. Such a system must be capable of producing so 
close a copy of the original that the copy will be indistinguishable 
from the original when both films are reproduced on identical equip- 
ment in the theater. The re-recording channel includes equalizing 
amplifiers and the necessary mixers. The latter are needed where 
additional sounds and speech effects are combined with the original 
pick-up recorded on the wide-range recording channel. 

April, 1934] WIDE-RANGE RECORDING 259 

Film processing assumes a r61e of even greater importance than 
heretofore since any degradation of quality produced by it is quite 
apparent in the impairment of naturalness or sound quality in the 
reproduced film. From this discussion it will be seen that the equip- 
ment for wide-range recording differs from that previously used prin- 
cipally in the use of the moving-coil microphone, the introduction of 
the new recording lens, the use of a low-pass filter, and the insertion 
of a dialog equalizer when required. Improved monitoring facilities 
are, of course, essential to its satisfactory use. 

As is usually the case when several such closely interrelated 
changes are to be made in an electrical or mechanical system, the at- 
tainment of a satisfactory over-all result requires that all the necessary 
modifications be properly coordinated and combined, so that the sys- 
tem is converted as a unit and not on a partial or piecemeal basis. 

In conclusion, it may be said that the conversion of a recording sys- 
tem to wide range results, first, in greater freedom of expression and 
action on the part of the actor and, second, in a much greater degree 
of naturalness and fidelity than has heretofore been achieved. This 
improvement is not only very apparent when wide-range recordings 
are reproduced in studio review rooms, but will also be readily ap- 
preciated by the layman in the theater. 


1 SNOW, W. B.: "Audible Frequency Ranges of Music, Speech, and Noise," 
/. Acoust. Soc. Amer., Ill (July, 1931), No. 1 (part 1), p. 155. 

2 WARD, J. S., AND WILLIS, F. C.: "Wide-Range Sound Reproduction," 
presented at the Fall, 1933, Meeting of the Soc. Mot Pict. Eng. 

3 WOLF, S. K.: "Acoustical Requirements for Wide-Range Reproduction," 
/. Soc. Mot. Pict. Eng., XXII (April, 1934), No. 4, p. 242. 

4 JONES, W. C., AND GILES, L. W.: "A Moving-Coil Microphone for High- 
Quality Reproduction," /. Soc. Mot. Pict. Eng., XVII (Dec., 1931), No. 6, p. 977. 

6 SIVIAN, L. J.: "Speech Power and Its Measurement," Bell Syst. Tech. J., 
VIII (Oct., 1929), No. 4, p. 646. 

6 SHEA, T. E., HERRIOT, W., AND GOEHNER, W. R.: "The Principles of the 
Light Valve," /. Soc. Mot. Pict. Eng., XVIII (June, 1932), No. 6, p. 697. 


G. E. ROTH* 

Summary. This paper provides further details of the Selenophon system described 
previously in the Journal, 1 and an account of recent developments of the system. 
Records can be made by either the variable-width or the variable-density method; the 
latter being used almost exclusively for film recording, and the former for recording on 
paper. The paper records were produced in order to provide a means of recording 
sound in the home and for other ''stenographic" purposes at small cost. 

The "Selenophon" system was developed as the result of many 
years of collaboration by the Director of the Institute of Theoretical 
Physics at the University of Vienna, Prof. Dr. Hans Thirring, the 
Director General of the Austrian Radio Broadcasting Company, 
Oscar Czeija, and the late Scientific Director of the Austrian Radio 
Broadcasting Company, Prof. Leopold Richtera. Schrott 1 has 
published a brief description of the process, and this paper provides 
further details concerning recent developments. Although the 
Selenophon system is of interest from the point of view of its re- 
cording process, the chief interest lies in the portability of the appa- 
ratus developed in recent years. 


For recording sound, the Selenophon process employs a string 
oscillograph, the optical arrangement of which is shown in Fig. 1. 
The light of a 50-watt incandescent lamp L, after passing through the 
condenser K and the slit-diaphragm B, passes through the reducing 
micro-objective to a fine metal wire S, about 0.1 mm. in diameter 
and 20 mm. in length, which is so stretched that its natural frequency 
lies above the highest frequency to be recorded. The metal wire 5 
is located in an air space 0.6 mm. wide between the poles of an electro- 
magnet; and is so arranged that, in its position of rest, it covers one- 
half the image of the slit B formed by 0. If, now, the microphone 

* Engineer, Tulpengasse 5, Vienna VIII, Austria. 



current flows through the wire, the wire is moved out of its position 
of rest and thus governs the intensity of the light passing the slit in 
correspondence with the sound vibrations to be recorded. Beyond 
the wire 5 is located a lens A, which images both the image of the 
slit and that of the wire together upon the film F, where the line 
of light has a breadth of 0.010 to 0.012 mm. The sound record and 
the focus on the film are observed by means of the focusing micro- 
scope composed of the objective O m and ocular A m . 

The maximum current impulses in the wire attain the value of 

FIG. 1. (Upper) Optical arrangement of the string oscillograph. 

FIG. 2. (Center) The relative positions of string and slit: (left) 
for the variable-width process; (right) for the variable-density 

FIG. 3. (Lower) The relative positions of string and slit: re- 
cording by the variable-width process. 

about 1 ampere; the resistance of the wire to alternating current 
amounts to about 0.5 ohm; the maximum power momentarily 
expended upon the wire accordingly attains values of 0.5 to 1 watt. 
In practice, a 4- to 5-watt microphone amplifier is used, in order to 
afford a reserve of output. The output transformer, which feeds into 
the oscillograph string and whose primary impedance is adapted to 
the output resistance of the terminal amplifier tube, possesses a con- 
version ratio of about 60 to 1, and has a secondary winding, in relation 



[J. S. M. P. E. 

to the current, of ample dimensions. The natural frequency of the 
string is 14,000 cycles per second. The use of oil for damping the 
string is avoided, since the oil bath absorbs the actinic rays and 
disturbances easily occur in the path of the rays as a result of eddy 
currents and occluded air bubbles caused by the rapid vibration 
of the string. A special damping method was developed, however, 
and is now employed; but it can not yet be described on account of 
the patent situation. The light source L and the electromagnet are 
energized by a 12-volt battery, and together require about 6 amperes. 

FIG. 4. View of recording apparatus. The driving motors are 
mounted on a base adjacent to the apparatus proper. Above is 
the speed indicator for the sychronous motor. The operator ob- 
serves the recording through the focusing microscope. His left 
hand rests on the lever that disengages the coupling between the 
motor and the camera. 

The recording can be done by either the variable-density or the 
variable-width method. The positions of the string with reference 
to the slit, corresponding to the two processes, are shown in Fig. 2. 

In recording on films, the Selenophon company uses at present 
the variable-density method almost exclusively; whereas for the 
paper records to be discussed later, recording is done only by the 
variable width-method, which is much less critical as to the average 

April, 1934] SELENOPHON SYSTEM 263 

recording illumination and the developing and printing processes. 
Fig. 3 (which is not drawn to scale) shows that the string is placed 
at a relatively small angle to the image of the slit. Such an arrange- 
ment has the advantage that a greater portion of the image is covered 
or uncovered by a small movement, d, of the string. In the re- 
cording apparatus, which is represented in Figs. 4, 5, and 6, the film 
is moved in a horizontal track which lies perpendicular to the optical 
axis of the image-forming system. Fig. 4 gives a general view of 
the apparatus, showing the sources of current and the driving motors. 
Fig. 5 is a view of the string galvanometer from the front, and Fig. 6 
shows the driving mechanism diagrammatically from above. The 

FIG. 5. String galvanometer, showing adjusting handles, connec- 
tions, and magnet field coils. 

driving is done through a shaft that extends parallel to the housing. 
This covers the supply sprocket V, the sound aperture where the 
exposure takes place, and the take-up sprocket N. V is driven by a 
pair of spiral gears Z b while TV is driven from the same axis through 
the pairs of spiral gears Z 3 and Z 4 . The sprocket T is driven from the 
main shaft through the spiral gears Z 2 , while the fly wheel S, which is 
rigidly joined to T, is rotated along with it by means of an elastic 
coupling. The film coming from the feed magazine K\ t after passing 
over the supply sprocket V, the sound gate F, the driving sprocket 



tf. S. M. P. E. 

T, and the take-up sprocket N, is taken up in the magazine K%, for 
which a separate motor M with a worm drive Z 5 is used. 

The sound recorder camera is driven by a d-c. motor of 250 to 300 
watts, held to a speed of 1440 rpm. by means of a small, four-pole 
synchronous motor (3-phase, 48 cycles). (Motors of similar type 
are used also for driving the picture cameras.) Since the main shaft 
of the sound camera should rotate at 180 rpm., a reduction gear unit 
having a ratio of 8:1 is used between the motor and the sound camera. 
This unit employs precision worm gears, and offers the advantage of 
permitting the motors to be mounted at right angles to the driving 
shaft. The transmission of vibrations from the separately mounted 
motors to the sound camera is thus largely avoided. A separable 
friction clutch is inserted between the drive and the sound camera, 
by means of which it is possible to run the motors up to their full speed 



_ ._ 

V F T N 

FIG. 6. The film propelling mechanism of the sound recording apparatus. 

and then, by slowly letting in the clutch, start the film in the sound 
camera gradually, avoiding the strain on the mechanism caused by 
sudden starting. Fig. 7 is a sectional view of the sound camera, 
the external appearance of which was shown in Fig. 5, through the 
optical system. The most important details are as follows : 

The recording lamp is at 15, in the lamp house 16. The foot 31 
of the lamp house can be adjusted in the clamping device 32 by 
loosening the screw 33, while the guides 34 and 35 permit a movement 
at right angles to the axis. The slit diaphragm 20 can be moved out 
of the path of the rays by the knurled head 21 fastened to the pin 22. 
Correct focusing of the point of light of the recording lamp in the 
optical axis of the condenser lens 19 is achieved by moving the parts 
17 and 18. The mounting of the lamp is designated by 30, while the 

April, 1934] SELENOPHON SYSTEM 265 

part 50 serves to adjust the slit diaphragm in a position exactly 
perpendicular to the axis of the film. The light rays, reflected per- 
pendicularly upward by the prism 23, pass through the hollow core 
24 of the string oscillograph 1, through the lens 25, through the hole 
26 in the magnet poles, and through the lens 27, and thus reach the 

FIG. 7. Diagrammatic cross-section through the sound recording camera, 
without the oscillograph string and its mounting. 

film 28, which is in the sound aperture 29. In Fig. 1, K corresponds 
to the condenser 19, B to the slit diaphragm 20, to the lens 25, 
and A to the lens 27. The lens 27 is focused by means of the disk 
36 with a milled edge; and the lens 25, by means of the disk 37. 
The parallel adjustment of the slit is accomplished by means of 38. 

266 G. E. ROTH \j. s. M. P. E. 

By means of the adjusting screw 53, the perpendicular optical axis 
of the string oscillograph can be displaced at right angles to the film 
width, the indicator 54 showing its position at any time. By such 
means, it is possible to make eight adjacent sound records, each 3 
mm. wide, on a 35-mm. film, with adjacent records running in op- 
posite directions. The arrangement of the string between the magnets 
is visible in Fig. 5. The string holder with its mounting can be turned 
about the principal axis of the oscillograph, this movement being 
governed by the projection 51 (Fig. 5). The output transformer of the 
amplifier is connected, on the one side, to the post 52, and is grounded 
to the apparatus on the other side at 43. 


A special opportunity of the Selenophon company lies in the con- 
struction of apparatus solely for sound film recording and repro- 

FIG. 8. Paper strip reproducing apparatus ready for use. 

ducing, to be used, like the phonograph, independently of motion 
pictures. The development of the small machines has already pro- 
gressed so far that they are used in regular broadcasting by all the 
Austrian stations, for international broadcasting (League of Nations' 
Session, September, 1931) and even by European and American 
stations. Two types have been developed: a small paper strip 
apparatus, which reproduces only sound records made on paper 
(Fig. 8) and a universal apparatus (7-7), shown in Figs. 9 and 10. 
Before going more into detail regarding these two very interesting 

April, 1934] SELENOPHON SYSTEM 267 

pieces of apparatus, something must be said about the use of "paper 
films." The Selenophon Company began with the proposition that 
the use of the usual celluloid film as a support for sound records in 
home apparatus could not be considered on account of the expense. 
There were two possibilities in the use of paper as a support: the 
prints can be made either by contact on photographic paper or by one 
of the known mechanical printing processes. Both possibilities have 
been developed to such an extent that they already fill high qualita- 
tive demands, so much so that the ground noise of the recording is 
substantially reduced in reproducing from paper by the construction 
of the illuminating system, and that the playing time of records on 
paper films is materially longer than that of disk phonograph records. 
The 300-meter reels, 6 mm. wide, carry two records each 2.5 mm. 

FIG. 9. The U-7 apparatus, closed, ready for recording. 

wide, and have a playing time of about 11 minutes per record; 
hence 22 min. in all. The sound range of the paper films printed 
on a rotary press corresponds to that of commercial disk records; 
their life is much longer, since the paper is subjected to no mechanical 

The small reproducing apparatus, shown in Fig. 10, is designed for 
6-mm. paper films and is sold as a suit-case model. It contains a 
built-in asynchronous driving motor, and a pre-amplifier that oper- 
ates on either 110 or 220 volts. The alternating" current de- 
livered by the pre-amplifier corresponds to the power from a good 
pick-up, and therefore operates satisfactorily radio receivers such as 
are found in any home. In Fig. 8, the path of the paper film is seen 

268 G. E. ROTH [j. s. M. P. E. 

clearly in the apparatus ready for operation; and, at the center, 
toward the back, the photocell is seen enclosed in a small cylindrical 

The universal apparatus U-7 permits both the recording and the 
reproduction of paper and celluloid films 6 mm. wide with two rec- 
ords. It was developed to comply with all the requirements of the 
radio broadcasting companies, which have, in the 7-7, an ideal 
sound recording apparatus that is relatively cheap (800 dollars). 
The sound is recorded directly on narrow film or light-sensitive paper 
strips without perforations. The records so produced are developed, 
fixed, washed, and dried in the usual manner, and can then be used 
again immediately for reproducing in the same apparatus. The 

FIG. 10. The U-7 apparatus arranged for reproduction of paper 
strip sound records. 

entire weight of the apparatus amounts to about 44 pounds. For 
transportation, a case of about 20 inches long, 20 inches high, and 12 
inches wide will serve. For operating the apparatus, only a two-tube 
amplifier is necessary, such as is found in radio receivers. Fig. 9 
shows the U-7 during recording. As will be seen, both the actual 
recording parts and the magazines are inclosed light-tight. The 
recording can be controlled optically and acoustically. The optical 
control is effected through a viewing microscope visible in the left- 
center of Fig. 9, while the acoustical control takes place through the 
photocell. In Fig. 9, the photocell housing is seen in the right- 
center with the shielded cable attached. At the base of the apparatus, 

April, 1934] SELENOPHON SYSTEM 269 

in the foreground, the lamp house and a part of the optical system 
for recording are visible. The 7-7 as a reproducing apparatus is 
shown in Fig. 10. In this case, the paper or celluloid strips are led 
through the same guide rolls as in the recording. Now, however, the 
sound lamp with its optical system serves as a light source for scan- 
ning the sound records. The photocell receives the light beam, 
which varies in its intensity after transmission or reflection, through 
the reproducing optical system. 


1 SCHROTT, P. VON: "The Selenophon Sound Recording System," /. Soc. 
Mot. Pict. Eng., XVHI (May, 1932), No. 5, p. 622. 


In the amendments of the Constitutions and By-Laws proposed at the Chicago 
convention last October, provision was made for five vice-presidents instead of the 
then existing two, and for an additional elective member of the Board of Govern- 
ors. The functions of the new officers were to direct the various agencies of the 
Society as indicated by the names assigned to the offices, indicated below. 

The following is a complete list of the officers and governors of the Society, in 
which the newly elected officers and governor are indicated by asterisks. 
President: Alfred N. Goldsmith 
^Executive Vice-President: H. C. Silent 
^Engineering Vice-President: L. A. Jones 
Technical Committees 

Sound Projection Theory 

Standards Projection Practice 

Studio Lighting Projection Screens 

Color Laboratory and Exchange Practice 

Non-Theatrical Equipment 
^Editorial Vice-President: J. I. Crabtree 
Board of Editors 
Papers Committee 
Progress Committee 
Historical Committee 
* Financial Vice-President: O. M. Glunt 
Membership Committee 
Ways and Means Committee 
Advertising Committee 

* 'Convention Vice-P resident: W. C. Kunzmann 
Convention Arrangements 
Apparatus Exhibit Committee 
Publicity Committee 
Treasurer: T. E. Shea 
Secretary: J. H. Kurlander 
Governors: H. T. Cowling 
*A. S. Dickinson 
R. E. Farnham 
H. Griffin 
W. B. Ray ton 
Chairmen of Local Section: 
Atlantic Coast Section: H. G. Tasker 
Mid- West Section: E. Cour 
Pacific Coast Section: E. Huse 


The Editorial and Convention Vice-Presidents were elected to serve for one 
year; the Engineering and Financial Vice-Presidents for two years. Future alter- 
nate elections of these officers will be for two-year terms. Of the Governors, the 
terms of Messrs. Cowling and Farnham expire January 1, 1935; those of Messrs. 
Dickinson, Griffin, and Ray ton, January 1, 1936. The terms of the President, 
Executive Vice-President, Secretary, and Treasurer expire January 1, 1935; those 
of the Section Chairmen, January 1, 1935. 


The second meeting of 1934 was convened at the General Service Studios, at 
Hollywood, February 21st, as a general session on "sound." The meeting was at- 
tended by 55 members and guests, whose interest in the papers was indicated by 
the spirited discussion following the presentations. 

Opening the meeting, Chairman E. Huse expressed the appreciation of the Sec- 
tion to the management of the General Service Studios for the use of their review 
room and wide-range reproducing equipment, and for the preparations and the 
courtesies extended to the Section. Following the reading and approval of 
the minutes of the previous meeting, Mr. H. C. Silent, Executive Vice-President 
of the Society, assumed the chair after words of appreciation by Mr. Huse for 
Mr. Silent's part in arranging for the two interesting papers presented by mem- 
bers of the Hollywood laboratory of Electrical Research Products, Inc. 

Mr. F. L. Hopper next presented a paper on "Wide-Range Recording on Film." 
Mr. Hopper's interesting and clear elucidation of the subject, accompanied by 
graphical and pictorial lantern slides illustrating the equipment and its character- 
istics, was followed by demonstration recordings and samples of production bear- 
ing upon important points of the discussion. 

Mr. D. T. Loye next presented a paper on the "Acoustics of Wide-Range Re- 
production." Characteristics of existing theaters and review rooms were con- 
trasted with those most desirable for the extended frequency range to attain the 
full benefit of the extension. The methods followed in making the acoustical 
measurements were explained, and a sound analyzing equipment was demon- 
strated in order to illustrate the important advances being made by equipment 
manufacturers in that field. 

After the presentations Mr. Huse resumed the chair. The members were then 
entertained by viewing one of the latest of the Silly Symphonies in color, which 
had been loaned for the occasion through the courtesy of the Walt Disney Studios; 
that it was greatly appreciated and well received need hardly be said. 

At the request of Mr. Huse, Mr. G. A. Chambers then proceeded to describe the 
composition and uses of the S. M. P. E. Standard Test Reels, after which the reels 
were reviewed on the screen. A lively and interested discussion of the various 
presentations and related subjects terminated the meeting. 


Members of the Society from eight cities of the Middle West met at Detroit, 
Mich., on March 3rd, to signalize the change of designation of the Section from 
"Chicago" to "Mid-West." The meeting convened at an afternoon session in the 


new studio of the Metropolitan Motion Picture Company, Mr. Maurice J. Caplan 
of that company kindly acting as host to the visiting members. 

After a brief summary of the aims of the S. M. P. E. by Mr. E. Cour, Chairman of 
the Section, Mr. H. L. Shippy of the Bausch & Lomb Optical Company presented 
a paper on the "Problems of Slide Film Projection." The remainder of the pro- 
gram of that session consisted of the following presentations : 

"The Slide Film vs. the Industrial Film," by Mr. John Strickler, of the Jam 
Handy Picture Service, Inc. 

"Little Smoke Screens," a 16-mm., optically reduced, sound-on-film Jam Handy 
production, projected by Mr. P. M. Albrecht, Davenport, la., with the new Victor 
16-mm. sound projector, and followed by a technical description of the new projec- 

"The Talking Slide Film," a demonstration by Mr. George Jarrett, of the 
Metropolitan Motion Picture Company. 

"Magnifying Time," by Mr. W. H. Strafford, a demonstration of the use of slow 
motion in engineering research at taking speeds of 400-1200 pictures per second. 

Steel and the Pierce- Arrow, a screening of two Metropolitan productions. 

At the conclusion of these presentations, the meeting adjourned for a visit 
through the interesting new plant of the Metropolitan Motion Picture Company, 
and after that, to dinner at the famous 2626. An interesting visit was then made 
to the plant of Wilding Picture Productions, Inc., where Mr. R. Biddy acted as 
host. The new wide-range recording equipment in process of being installed in 
the Wilding Studio was very thoroughly explained to the members by Mr. E. A. 

The evening session was held in the studios of Jam Handy Picture Service, Inc. 
Mr. Tex Rickard demonstrated the ERPI sound recording equipment, after which 
the members were conducted on a tour of inspection through the three large 
plants of the Jam Handy Company by Messrs. G. Knapp and J. F. Strickler. 
The tour ended in the projection room, where Love Apples and Men and Work 
were screened, after which the session adjourned. 



W. C. KUNZMANN, Chairman 



H. BLUMBERG, Chairman 





H. GRIFFIN, Chairman 


Officer and Members of Atlantic City Local No. 310, 1. A. T. S. E. 


MRS. M. C. BATSEL, Hostess 

Assisted by 




The Convention will convene at 10:00 A.M., Monday, April 23rd, at the Chal- 
fonte-Haddon Hall, in the Viking Room on the thirteenth floor of the Haddon Hall 
section. At noon of the opening day there will be an informal get-together 
luncheon, during which the members of the Society will be addressed by several 
prominent speakers. The morning preceding the luncheon will be devoted to 
registration, reports of officers, and other Society business, as well as the reports 
of technical committees. 


All technical sessions and film exhibitions will be held in the Viking Room, 
where also will be located the registration headquarters. Technical sessions 
will be held on Monday, Tuesday, and Thursday afternoons, and on Tuesday, 


274 SPRING CONVENTION [j. s. M. P. E. 

Wednesday, and Thursday mornings. Monday morning will be devoted to 
Society business and committee reports; Wednesday afternoon, preceding the 
semi-annual banquet in the evening, will be left free for recreation. The film 
programs of recently produced outstanding features and shorts will be held on 
Monday and Tuesday evenings, and will be booked by Mr. J. Greenburg, of 
the Philadelphia Film Board of Trade, and Mr. H. Blumberg, chairman of the 
Local Arrangements Committee. 


The S. M. P. E. Semi-Annual Banquet and Dance will be held in the Rutland 
Room of the Chalfonte-Haddon Hall on Wednesday, April 25th, at 7:30 P.M. 
an evening of dancing, movies, and entertainment; no banquet speeches. Ban- 
quet tickets should be obtained at the registration headquarters; tables reserved 
for six or eight persons. 


Excellent accommodations are assured by the management of the hotel, and 
minimum rates are guaranteed. Room reservation cards mailed to the member- 
ship of the Society should be returned immediately to the Chalfonte-Haddon 
Hall in order to be assured of satisfactory reservations. 


Room with bath, ocean view, single $4.00 
Room with bath, ocean view, double $6.00 
Room with bath, city view, single $3.00 
Room with bath, city view, double $5.00 


A reception suite will be provided for the use of the ladies attending the Con- 
vention, and an attractive program for their entertainment is being prepared by 
the Ladies' Committee. 


Arrangements are being made to hold an exhibit of newly developed motion 
picture apparatus, in order to acquaint the members of the Society with the 
newly devised tools of the industry. The exhibit will not be of the same nature 
as the usual trade exhibit; there will be no booths, but each exhibitor will be 
allotted definite space and all exhibits will be arranged in a single large room. 
Requests for space should be directed to the General Office of the Society, 33 
West 42nd Street, New York, N. Y., stating the number and nature of the items 
to be exhibited. The charges for space will be as follows: up to 20 sq. ft., $10; 
every additional 10 sq. ft., $5. 

April. 1934] SPRING CONVENTION 275 

Monday, April 23rd 

9 : 00 A . M . Viking Room 


Society Business 

Reports of Officers 

Reports of Committees 
12:30 P.M. Benjamin West Room 

Informal get-together luncheon for members and guests; short ad- 
dresses by prominent speakers 
2 : 00 P.M. Viking Room 

Program of Technical Papers 
8 : 00 P.M. Viking Room 

Presentation of recent outstanding motion pictures 

Tuesday, April 24th 

10:00 A.M. Viking Room 

Program of Technical Papers 
2 : 00 P.M. Viking Room 

Program of Technical Papers 
8 : 00 P.M. Viking Room 

Presentation of recent outstanding motion pictures 

Wednesday, April 25th 

10 : 00 A.M. Viking Room 

Program of Technical Papers 
2:00 P.M. Afternoon open for recreation 
7:30 P.M. Rutland Room 

Semi-Annual Banquet of the S. M. P. E.; an evening of music, 
dancing, entertainment, and motion pictures 

Thursday, April 26th 

10:00 A.M. Viking Room 

Program of Technical Papers 
2: 00 P.M. Viking Room 

Program of Technical Papers 
5:00 P.M. Adjournment of Convention 



Prepared under the Supervision 



Two reels, each approximately 500 feet long, of specially pre- 
pared film, designed to be used as a precision instrument in 
theaters, review rooms, exchanges, laboratories, and the like 
for testing the performance of projectors. The visual section 
includes special targets with the aid of which travel-ghost, 
lens aberration, definition, and film weave may be detected 
and corrected. The sound section includes recordings of 
various kinds of music and voice, in addition to constant 
frequency, constant amplitude recordings which may be used 
for testing the quality of reproduction, the frequency range 
of the reproducer, the presence of flutter and 60-cycle or 96- 
cycle modulation, and the adjustment of the sound track. 


(Shipped to any point in the United States) 

Address the 





Volume XXII MAY, 1934 Number 5 



Two New Photographic Recording Instruments, the Sensito- 

graph and the Gammagraph . . H. BRANDES AND R. SCHMIDT 279 

A Study of Television Image Characteristics . . E. W. ENGSTROM 290 

Transmission and Reproduction of Speech and Music in Audi- 
tory Perspective . . . H. FLETCHER 314 

Book Reviews 330 

Society Announcements 331 

Atlantic City, N. J., Convention; April 23-26, inclusive 334 





Board of Editors 
J. I. CRABTREE, Chairman 



Subscription to non-members, $8.00 per annum; to members, $5.00 per annum, 
included in their annual membership dues; single copies, $1.00. A discount 
on subscriptions or single copies of 15 per cent is allowed to accredited agencies. 
Order from the Society of Motion Picture Engineers, Inc., 20th and Northampton 
Sts., Easton, Pa., or Hotel Pennsylvania, New York, N. Y. 

Published monthly at Easton, Pa., by the Society of Motion Picture Engineers. 

Publication Office, 20th & Northampton Sts., Easton, Pa. 
General and Editorial Office, Hotel Pennsylvania, New York, N. Y. 
Entered as second class matter January 15, 1930, at the Post Office at Easton, 
Pa., under the Act of March 3, 1879. Copyrighted, 1934, by the Society of 
Motion Picture Engineers, Inc. 

Papers appearing in this Journal may be reprinted, abstracted, or abridged 
provided credit is given to the Journal of the Society of Motion Picture Engineers 
and to the author, or authors, of the papers in question. Exact reference as to 
the volume, number, and page of the Journal must be given. The Society is 
not responsible for statements made by authors. 

Officers of the Society 

President: ALFRED N. GOLDSMITH, 444 Madison Ave., New York, N. Y. 
Executive Vice-President: HAROLD C. SILENT, 7046 Hollywood Blvd., Los 

Angeles, Calif. 

Engineering Vice-P resident: LOYD A. JONES, Kodak Park, Rochester, N. Y. 
Editorial Vice-President: JOHN I. CRABTREE, Kodak Park, Rochester, N. Y. 
Financial Vice-President: OMER M. GLUNT, 463 West St., New York, N. Y. 
Convention Vice-President: WILLIAM C. KUNZMANN, Box 400, Cleveland, Ohio. 
Secretary: JOHN H. KURLANDER, 2 Clearfield Ave., Bloomfield, N. J. 
Treasurer: TIMOTHY E. SHEA, 463 West St., New York, N. Y. 


EUGENE COUR, 1029 S. Wabash Ave., Chicago, 111. 
HERFORD T. COWLING, 7510 N. Ashland Ave., Chicago, 111. 
ARTHUR S. DICKINSON, 28 W. 44th St., New York, N. Y. 
RALPH E. FARNHAM, Nela Park, Cleveland, Ohio. 
HERBERT GRIFFIN, 90 Gold St., New York, N. Y. 
EMERY HUSE, 6706 Santa Monica Blvd., Hollywood, Calif. 
WILBUR B. RAYTON, 635 St. Paul St., Rochester, N. Y. 
HOMER G. TASKER, 41-39 38th St., Long Island City, N. Y. 




Summary. -Two pieces of apparatus for evaluating sensitometric strips are 
described: the sensitograph for objective measurement of complete sensitometric 
strips and automatic recording of the density curve, and the gammagraph for use in 
continuous control of constancy of the gamma value of short strips. 

A photographic density curve is a .curve which represents the 
relation between exposure (plotted logarithmically) and the density 
produced by it. To produce the curve, strips of the test material 
are so exposed in a sensitometer that the amount of light falling on 
consecutive steps of the layer increases in geometric progression. 
The strips are developed and then measured step by step in a suitable 
densitometer, and the results are plotted graphically. The measuring 
and the plotting done at the present time wherever large numbers 
of plant tests have to be made, as in photographic factories, require 
considerable time. 

Moreover, there is the disadvantage of subjective error with 
measuring instruments, so liable to occur in a long continued series 
ot measurements, because the observer's eye quickly tires, and errors 
of measurement result. A certain saving of time can be produced 
by combining the measuring and the drawing in one operation (as 
in the Goldberg densograph of Zeiss Ikon) but the saving of time is 
not very great and, in addition, the measurements are made with the 
eye, and are subject to the error discussed above. Our aim, therefore, 
was to develop a device that would make the measurements ob- 
jectively, and reduce as far as possible the time required ior measuring 
the sensitometric strips. 

For that purpose two instruments were developed which combine 
the following advantages: (1) objective measurement of the den- 

* Translated from Agfa Veroffentlichungen des wissenschaftlichen Zentral- 
Laboratoriums, der Photographischen Abteilung, Band III, 106-114. 




[J. S. M. P. E. 

sities, by using a photoelectric cell instead of the eye; (2) substantial 
saving of time by making the measuring and plotting process com- 
pletely automatic. The range of application of the two instruments 
is adjusted to fit the particular problem : the sensitograph is used for 
evaluating the entire sensitometric strip; that is, for plotting the 
whole density curve, while in those cases in which only the main- 
tenance of constancy is concerned, e. g., the developing conditions in 
processing stations, the use of a shortened sensitometric strip and 
its evaluation by the gammagraph is in order. 


(.4) Principles. A density wedge is arranged between a light 
source and a photoelectric cell which can be moved by means of a 

FIG. 1. Principle of the sensitograph. 

driving arrangement in the sense of increasing the illumination 
falling upon the photoelectric cell. When the intensity of the illumi- 
nation attains a certain value, the clutch between the drive and the 
density wedge is disengaged, and the wedge remains stationary (null 
point). When a density strip to be measured is introduced into the 
path of the rays, then the wedge is shifted until the sum of the density 
of the wedge and the field being measured is equal to the density of 
the wedge at the null point (measuring process). The position of the 
wedge when a balance is attained is thus a direct measure of the 



density of the field. In order to render the fluctuations of the light 
source ineffective, a second photoelectric cell (compensating cell), 
which receives its light from the same incandescent lamp as the 
measuring cell, is used in place of the usual high resistance in the 
measuring cell circuit. 

The method of plotting the results is shown schematically in Fig. 1 
in which we see the light source (12), the measuring cell (8), and the 
compensating cell (9). The sensitometric strip and the measuring 
wedge (7), fastened to the carrier (28), lie in the path of light to the 
photoelectric cell (8). The wedge is driven through the clutch (13) 

FIG. 2. Optical arrangement of the sensitograph. 

and the steel tape (30) to which the recording point (29) is attached. 
With increasing density of the sample the measuring wedge is moved 
farther and farther to the left and the pen is moved toward the top 
of the graph paper, increasing correspondingly the ordinates of the 
points to be registered. The form itself is fastened to a carriage (25), 
which runs on the rails (53), and also carries the sensitometric strip 
holder (28). This carriage is moved to the left step by step in such a 
way that each area to be measured is brought into position when its 
corresponding abscissa value is brought under the pen. In this way 
the density curve is plotted by combining the movement of the 
carriage and the recording point, which operate perpendicularly 
to each other. 

(B) Optical Arrangement. Fig. 2 shows the lamp house (12) with 
the photoelectric cells (8 and 9), as well as the sensitometric strip 



[J. S. M. P. E. 

(6) and the wedge (7), on a larger scale. A 500- watt motion picture 
projection lamp (1), with the condensers (2 and 3) and the filter 
cell containing copper sulfate solution (4) used to prevent heating, 
provides the illumination. The diaphragm (5) reduces the beam 
to the width of the sensitometric field. The compensation cell (9) 
receives its light by reflection from the white screen (10). There is 
provision for inserting a density at (11) for setting the zero point. 
The housing and trough are provided with a water jacket. 

(C) Electrical Arrangement. As soon as the illumination of the 
photoelectric cell (8), or, more accurately, the relation of the illumi- 

FIG. 3. Circuit diagram of the sensitograph. 

nation of the cell (8) to the illumination of cell (9), falls below a 
certain value, the clutch (13), which controls the gray wedge (7), is 
engaged; and then, upon the restoration of that value, is disengaged. 
The electrical circuit, shown in Fig. 3, serves this purpose. If, with 
constant light illumination of the photoelectric cell (9), the illumina- 
tion of the photoelectric cell (8) decreases, then obviously the anode 
current of the tube (14) increases, with a consequent increase in the 
potential across the resistance (15), and the anode current of tube (16) 
is diminished. Thus, with less than the standard illumination, the 
relay (17) is released, closing the circuit that serves the magnet (18). 



Since the current strengths at disposal for controlling the grid 
potential of the first tube are extremely low (of the order of 10~ 8 
ampere) on account of the low level of illumination of the photo- 
electric cell, extensive protective precautions are necessary to mini- 
mize leakage. They consist of grounding the cathode of the first 
tube so that there is essentially no potential difference across the 
insulation between the grid and ground. The only leakage currents 
still to be feared would be on the part of the cathode lead of the 

FIG. 4. Mechanical arrangement of the sensitograph. 

cell (8), the anode lead of the cell (9), and the amplifier tubes. The 
leads of the photoelectric cells in question were insulated by leading 
them through a grounded brass ring, which makes contact with 
the hard rubber mounting plate of the cell and provides a path to 
ground for leakage currents coming from the cell leads to the ground. 
Grounded wire rings, wound around the cells at suitable places, 
screen off disturbing potentials. The anode lead at the base of the 
amplifier tube (14) is screened by means of a grounded tinfoil ring. 
The pin connected to the grid at the foot of the tube was not, as 
customarily, plugged into the socket, but after removal of the corre- 

284 H. BRANDES AND R. SCHMIDT [j. s. M. p. E. 

spending socket part wired directly. A degree of insulation was 
attained by these precautions which made it possible to record 
densities reliably up to D = 3.5, using the illumination intensity 

(D) Mechanical Arrangement. The mechanical arrangement 
(Figs. 4 and 5) fulfills the following functions : 

(1) To move the strip to be measured forward step by step. 

(2) To move forward, simultaneously, the record sheet under the pen, in 
steps corresponding to the single steps of the strip. 

(3) To bring about the movement of the control for the gray wedge and the 
recording point through the action of the magnet (18). 

(4) To operate the recording point. 

FIG. 5. The recording device. 

A 65-watt motor (19) is used as the source of power for all purposes. 
The screw (20) drives the shaft (21), which in turn actuates the 
various elements in the machine. 

For (1) and (2) above: the shaft (21) drives the shaft (22) by means 
of a six-sectioned Maltese cross driving gear. The pinion (23) on 
this same shaft drives the rack (24), which is attached to the carriage 
(25). The carriage is thus moved once for every rotation of (21). 
It propels both the graph paper (27) and the carrier (28), to which 
the sensitometric strip is attached. 

For 3: the gray wedge (7) and the pen (29) are both fastened to 
the endless steel band which is carried over the four rollers (31) and 
passes in front of friction roller (32) which is driven from shaft (21) 



by bevel gears. As soon as the magnet (IS) is energized it presses 
the steel band against the driving roller by means of the lever (33) 
and the idle roller (34). The steel band, together with the gray 
wedge and the recording point, are moved until the correct standard 
illumination of the photoelectric cell (8) is attained, and the magnet 
(18) is cut out. At once the spring (35) retracts the lever (33), and 
the steel band is no longer moved but is held fast by the brake 
block (36). In this way the adjustment of the recording point is 


1 2 

3 4 

6 7 

8 9 

I I 

3.0 3,5 +,0 4.5 
10 11 12 13 14 15 

I I 

5.0 5.5 6.0 
16 17 18 19 20 

FIG. 6. Curve sheet with plotted points. 

The location of the pen is established by the point at which the 
coupling is switched out. There is the danger that with very small 
increases in density from step to step on the sensitometric strip, 
such as occur in the neighborhood of the threshold, the darkening 
of the photoelectric cell that occurs is not sufficient to actuate the 
relay for engaging the clutch. In order to avoid this, an arrange- 
ment is provided that draws the density wedge and the recording 
point backward whenever any advancement of the carriage takes 
place that is sufficient to cause the relay to act. The measuring 
device is thus set anew for each field, even when no increase in density 
over the preceding field occurs. The arrangement provided for the 


purpose consists of a dog (46), which is fastened to the disk (21) and 
moves the lever (47) to the right with every rotation of the disk, in 
such a way as to allow it to go back again quickly. When the lever 
is drawn back first the contact (48) which breaks the circuit of the 
magnet (18), and then the claws (49) grasping the carrying band 
(30) close. With further movement of the lever (47) the entire 
lever (50), together with the claw (49) and the steel band (30), is 
moved to the left until the dog (46) releases the lever (47), where- 
upon all quickly return to the rest position under pressure of the 

FIG. 7. Interior view of the gamma- 

springs (51 and 52). The steel band nevertheless remains displaced 
about 4 mm. 

For 4: the recording stroke occurs when the bar (37) is lowered. 
This is brought about by a cam on the shaft (21) acting upon a roller 
affixed to the end of a lever (41), which causes the bar to be raised 
and then dropped shortly before the next advancement of the carriage. 
The bar (37) thrusts the spring-supported marker (38, Fig. 5) toward 
the graph paper, and the marking is done by means of the type- 
writer ribbon (40). The marker is a rotatable hexagon which carries 



six different kinds of type (cross, point, circle, etc.) so that several 
curves can be recorded differently on the same graph paper for 
purposes of comparison. 

The single phases of measurement take place in the following 
order: advancement of the density step and the graph paper; 
adjustment of the recording point; recording; advancement, etc. 
With the last advancement the carriage cuts out the switches (43) 
and (44), that is, the motor and the lamps, by means of the trip (42). 

The steps of the sensitometric strips are 10 mm. wide. Accord- 


FIG. 8. Circuit diagram of the gammagraph. 

ingly, the advancement of the carriage with the recording paper 
must be adjusted to 10 mm. per step, which requires the constant 
of the gray wedge to be 0.3 per cm. A curve sheet with two curves 
is shown in Fig. 6. 


The gammagraph had its origin in the development of the Agfa 
gammameter. The latter consists of a printing sensitometer in 
which a five-step sensitometric strip is exposed, and a reading desk on 
which one can obtain the gamma value of the finished developed 
strips. The gammagraph, which draws the results as a series of 
points on a recording strip, takes the place of the reading desk. In 
this case, the problem is limited for most purposes to indicating 



[J. S. M. P. E. 

whether the gamma value of the strip to be tested is correct, or lies 
too high or too low in comparison with a desired value. 

For this purpose the test strip is laid on a comparison strip of the 
desired gamma, and is tested photoelectrically as to whether the total 
transmission is the same in all five fields or whether it increases 
or decreases from step to step. 

Fig. 7 shows a cylindrical sector (7) rotatable about the axis (2) 
which contains the comparison strip and is arranged for taking the 
test strip. It carries also a comparison field (3) with the help of which 
the absolute value of the density of the gammameter strips can be 
compared. During the rotation of the sector by means of the clock 

FIG. 9. The gammagraph ready for use. 

work (4) all six fields are moved through the functioning of the cam 
(5), one after the other, in front of the window of the lamp house 
(6). The light passing through the gammameter strip and the com- 
parison strip falls on the photoelectric cell (7) which, through the 
agency of the amplifier tube (#), actuates the galvanometer (9). 
The position of the galvanometer indicator for each area is marked 
on a recording tape running over the roller (12) by the dropping of 
the frame (10) and the typewriter ribbon (11). The drive for the 
recording paper comes from the rubber roll (13) and the pressure 
roller (14), and is usually actuated by clockwork, as is also the con- 
trol of the dropping frame. 

The circuit diagram of the apparatus is shown in Fig. 8, where I 
is the measuring apparatus, and // an auxiliary instrument that 


serves to connect the main part with a 110- volt source of continuous 
current. The method of working is shown without going any further 
into the explanation of the circuit. The figures correspond to those 
in the photograph. The regulating resistance (75) serves to maintain 
the working potential, read at the voltmeter (16), constant when the 
supply voltage varies. The switch (17) controls the entire apparatus. 
Fig. 9 shows the gammagraph hung up ready for use. The sample 
strip is inserted in the cylinder (1) through the opening (18), and the 
current is turned on by means of the switch (17). Then the button 
(19) is pressed, setting the clockwork in action. The apparatus 
automatically takes care of five measuring fields and the com- 
parison field, indicates the results on the recording paper, and 
after the measurement is done automatically cuts out the clockwork 



Summary. An investigation was carried out to obtain quantitative information 
on the several characteristics of television images, particularly those relating to image 
detail. The tests were conducted largely through the use of equivalents so as to pro- 
vide sufficient range of measurement. Such data are of value in establishing op- 
erating standards, determining satisfactory performance, and in guiding development 
work. It was found possible to define satisfactory television image characteristics 
for those items studied. The results are given in such form as to be readily applicable 

to practical conditions. 


Because of the lack of quantitative measures of performance, ex- 
pression of the degree of satisfaction provided by a television image 
has been bounded on one hand by the optimism or conservatism of the 
observer, and on the other hand by the practical limitations which 
prevent for the moment an increase of picture detail, picture steadi- 
ness, picture illumination, picture contrast, and frame repetition fre- 
quency. It is the purpose of this paper to describe investigations 
made regarding some of these picture properties. 

Picture detail is determined by the quantity of information that 
the entire system can handle in a given time. Also, the communica- 
tion band is proportional to the frame repetition frequency. (Frame 
repetition frequency determines steadiness of action and picture 
flicker.) Optical, sensitivity, and transformation problems are 
present in the pick-up gear and become apparent as attempts are made 
to go beyond present practical limits. Somewhat similar problems 
are present in the reproducing elements. These limits are contin- 
gent upon the particular state of the art, and, therefore, are con- 
stantly receding and yielding to development. 

Since the frequency band required is proportional to the quantity 
of information to be transmitted, the limitations of the electrical 
channels must be considered. These problems include the ability to 
handle wide frequency bands and to provide space in the radio spec- 

* Reprinted from Proc. I. R. E., 21 (Dec., 1933), No. 12, p. 1631. 
** RCA Victor Co., Camden, N. J. 


trum for television channels. This may be illustrated by the follow- 
ing table for certain conditions which are stated. 

Aspect Ratio 1.33 (4/3). 

Frame Repetition Frequency 24 per second. 

Picture Frequency 

It is assumed that the picture resolution along the scanning line is ap- 
proximately the same as the width of the scanning line (square picture elements) 
and that each picture element (of maximum resolution) requires one-half cycle for 
transmission in elemental form. The maximum picture frequency, therefore, 
determines the steepness of wave front or change in contrast along the scanning 

It is also assumed that pictures will be transmitted for ninety per cent of the 
total time, the remaining ten per cent being necessary for control functions. 





























The limitations present in the electrical circuit are also determined 
by the state of the art at any particular time, and, therefore, are sub- 
ject to advances as a result of development. It is probable that the 
ultimate limit may be the space available for television channels in the 
radio spectrum. 


Determination of satisfactory picture quality in television images is 
difficult because of the inadequacy of present television apparatus for 
such a study and because the reactions involved are largely psycho- 
logical and physiological. During the growth of television detail, as 
the development work has progressed, improvement of picture quality 
has been noted, for example, through stages of a, 2a, 3a, and 4a scan- 
ning lines, where 4a represents the present practical limits. We are 
not in a position to work with and study 5&, 6a, etc., scanning lines in 
such a determination. Therefore, in studies of picture detail, pic- 
ture size, and viewing distance, many subterfuges have been used. 

Because of the wealth of detail, extreme ranges of brightness, and 
contrast in nature, the eye tends to demand image resolution up to the 

292 E. W. ENGSTROM [j. a M. p. E. 

acuity and perception limits of the eye. We have, however, become 
accustomed to certain compromises in these image characteristics 
through long association with paintings, photographs, projected trans- 
parencies, and other forms of reproduction, because of the limitations 
of these agencies of reproduction. 

The perception of form or acuity of the eye is usually defined as the 
minimum angular separation which permits resolution of two point 
objects. For the average normal eye this approximates one minute 
of arc for that portion of the field which falls on the f ovea of the retina. 
Other measures include minimum dimensions for seeing a point, line, 



FIG. 1. Patterns for visual acuity tests. 

or separation between two lines or groups of lines, change of contour, 
etc. Some of these become rather indefinite if the object is self-lumi- 
nous. Other eye characteristics of interest in such a study include per- 
ception of movement, perception of contrast, color vision, color sensi- 
tivity, perception of light, and effects of flicker. 

Elementary studies of some properties of vision may be made 
through the use of the chart indicated by Fig. 1. This chart includes 
a group of patterns which may be obtained from the scanning system 
used in television. The numbers under each group indicate the total 
number of scanning lines for the height of the chart. This chart as- 
sumes equal horizontal and vertical resolution for the groups of five 

May, 1934] 



figures to the left of the chart. It also assumes that the scanning lines 
will coincide with the detail structure (of same width as scanning line) 
of the chart scanned, so as to provide the greatest possible detail in the 
chart reproduced for a given number of scanning lines. The fine grat- 
ing to the right of the chart indicates the scanning line paths. No 
particular attempt was made to avoid optical illusions, but it is be- 
lieved that the figures are sufficiently free to avoid mistakes in judg- 

Relationship of picture size, picture detail, and viewing distance is 
of interest in studying television images. This relationship may be 
approached from theoretical considerations of providing sufficient de- 
tail to satisfy the acuity of the eye. We may start with the definition 
of acuity for the average normal eye (one minute of arc for that por- 
tion of the field which falls on 
the f ovea of the retina) . This 
is justified even though the 
image may be so large as not 
to be included within the 
relatively small field of most 
acute vision, since the eye 
naturally tends to explore the 
entire image, and the image 
is, therefore, subjected in all 
its parts to the finest resolu- 
tion of the eye. 

Since the resolving proper- 
ties of the eye are so definitely 
tied up with the type of de- 
tail to be analyzed, we shall choose for this theoretical considera- 
tion a very specific definition of acuity. For this example we 
shall use two black lines separated by a white space equivalent 
to the width of one of the lines, such as the pairs of lines in the 
groups to the left of the chart. If such lines, for a particular viewing 
point, are separated so that the distance between them subtends an 
angle to the eye of one minute, then the average eye will be able to see 
them as two lines. At greater viewing distances the two lines will blur 
into one. In order to keep our discussion in terms of scanning lines, 
the curve to follow will be plotted in terms of scanning lines against 
viewing distances. For the two horizontal lines it is necessary to have 
one scanning line for each line and one scanning line for the space be- 



FIG. 2. Scanning lines per inch required 
for subtending various angles at the eye for 
various viewing distances. 



[J. S. M. p. E. 

tween lines. Since, by definition, the space between lines must sub- 
tend an^angle of one minute, then the width of each scanning line or, 
in other words, the distance between centers of scanning lines, also 
subtends an angle of one minute. Fig. 2 includes a calculated curve 
indicating for various viewing distances the number of scanning lines 
per inch required for a one minute of arc separation between centers of 
scanning lines. For later reference, curves are also shown for one- 
half-minute and two-minute arc distances between centers of scanning 

In using these curves it is necessary to understand the span or varia- 
tion of eye acuity for different people. For the type of detail we are 
considering, this span is probably from approximately one-half minute 

to approximately one and 
one-half minutes one minute 
being used as the average. 
This is pointed out specifi- 
cally because of this wide 
variation and the difficulty 
of dealing with a definite 
average value. 

By inspection of this curve 
(the one-minute curve) we 
are able to determine (within 
the scope of our definition) 
the amount of detail in terms 
of scanning lines for still im- 
ages at various viewing dis- 
tances for the "average eye." If, for viewing distance X, the curve 
indicates that Y scanning lines should be provided, then the eye will be 
satisfied at this viewing distance for a detail of Y scanning lines. For 
closer viewing distances and Y scanning lines, the eye will not be 
satisfied since the picture structure will be pronounced, resulting in 
"lack of detail." For greater viewing distances and Y scanning lines, 
the eye will be satisfied from the standpoint of detail, but more detail 
is available than required by eye acuity. 

In order to make some practical tests, a number of observations 
were made using charts of the type shown in Fig. 1. Three charts 
were used one two and one-half inches high, the second five inches 
high, and the third twenty inches high so as to provide an effective 
range of scanning lines of from 60 to 480. Tests were made by three 







Two .Lines 
Two Squares 
Crossed Lines 










\ S 








D \ 













k "^ 

* . 







2 4 6 8 10 18 14 16 18 20 28 84 

FIG. 3. 


Observational curves obtained 
for several patterns. 


people having good vision and having no known eye defects. The 
tests were conducted by placing the charts on a wall at eye level in a 
room having uniform daylight illumination (mainly sky light since the 
sun did not strike the windows). The illumination at the chart was 
between 20 and 40 foot-candles. The contrast on the charts was the 
maximum possible in a normal photographic print. 

The pairs of lines to the left of the chart were used to obtain data 
for the first curve. For each degree of "scanning line detail" a view- 
ing distance was chosen at which the two lines could just be resolved; 
at greater distances the two lines blurred into one. At this same 
viewing distance the group of horizontal and the group of vertical 
lines (the second and the third groups of figures from the left) could 
just be resolved into lines; at greater distances they blurred into a 
uniform gray. The curve plotted in Fig. 3 is the average for the three 
observers. A curve was plotted for resolving the two squares, a part 
of the fourth group from the left (the two squares at the left, just 
above the checkerboard pattern) . A curve was plotted for resolving 
the checkerboard pattern, in the lower half of the fourth figure from 
the left. A curve was also plotted for the crossed lines, the fifth 
figure from the left. In this case the viewing distance chosen was 
the point at which the line structure could just be seen; at greater 
viewing distances the line structure was missing, and the cross 
appeared to be made up of two straight lines of constant width. All 
of the curves were plotted using the average viewing distance for the 
three observers. An interesting point noted from the observations 
was the consistency of the viewing distances chosen. Two of the 
observers picked viewing distances very nearly the same. The 
third observer picked viewing distances slightly greater (10 to 20 per 
cent). In the case of the third observer, this difference was con- 
sistent for all of the tests. These curves indicate the range of satis- 
factory viewing distances for the types of detail chosen. In general, 
the detail types probably do not cover the extremes, but do cover at 
least the average range encountered in scanned television images. 
Some interesting deductions may be made by comparing these data 
with the theoretical curves for one-half, one, and two minutes of arc 
separations between center lines of scanning paths. For convenience 
these curves are shown in Fig. 4, superimposed. The data from the 
observations are indicated by a dark band including the span be- 
tween the test for the two lines and the test for the crossed lines of the 
previous curves. The one-half-, one-, and two-minute arc curves are 



[J S. M. P. E. 

shown as solid lines. The data presented in Fig. 4 indicate that the 
types of detail on which the tests were made require, for any chosen 
viewing distance, a range of from a little over one-half minute to a 
little less than two minutes of arc separation between centers of 
scanning lines. It is also indicated that the average acuity of the 
three observers is above that of the "average eye" near the upper 
limit of acuity. From the standpoint of these tests and the tests to 
follow, this is a safe condition because, for any viewing distance, de- 
tail satisfactory to this group of three observers will certainly be 
satisfactory to the average observer. 


FIG. 4. Comparison of observational data and curves 
of Fig. 2. 

In viewing reproductions the observer tends to position himself so 
that he is satisfied regarding the information and the effect he wishes 
to obtain. (The position or viewing distance for greatest resolution 
is about eight to ten inches for the average person.) Because of habit 
and experience we have learned to temper our acuity demands. The 
following generalizations are of interest, and are given in terms of 
general experience rather than technical knowledge. When viewing 
a painting we rather unconsciously choose a position where the brush 
stroke detail becomes unnoticeable, and where we obtain the effect the 
artist wished to convey. We have learned that a newspaper illus- 
tration contains only a certain amount of detail, and that such illus- 
trations will not bear close inspection. We also know in general what 


to expect from motion pictures of the theater and home types. We, 
further, know that good photographs go beyond the acuity limits of 
the eye, and that the field may be optically enlarged to improve the 
resolution. Other examples could be given, but the above are suf- 
ficient to illustrate the effect of experience on the average person. 

The value of the above curves is to indicate the maximum useful de- 
tail from the standpoint of eye acuity, assuming favorable conditions 
for all other related factors. For a particular viewing distance, the 
amount of detail required in a reproduction (still image) is dependent 
upon the type of information to be conveyed by the picture. Since 
this varies, it is safe to assume that, for limiting conditions, detail cor- 
responding to that indicated by the curves should be provided. For 
average conditions and for general use it is also safe to assume that 
sufficient satisfaction can be provided by considerably less detail than 
that indicated. This is verified by the various types of printed re- 


It is difficult to interpret television image quality in terms of the 
relationships discussed. The first reason for this is that television 
images are the result of scanning at the pick-up end which introduces 
an aperture effect, and at the reproducing end the aperture effect is 
introduced for the second time. This results in a definite and peculiar 
line and detail structure. Detail along each line is dependent upon the 
ability of the system to reproduce changes in contrast. The second 
reason is that television images are made up of rapidly superimposed, 
individual pictures much the same as motion pictures. The third 
reason is that television images usually include motion having certain 
continuity. The effects of motion will be taken up more in detail 
later in the paper. 

Photographs have been made which consist of scanned reproduc- 
tions of an ordinary photograph or scene. These, therefore, have 
picture structures which correspond to television images and are 
useful in studies of the character outlined by this paper. Such 
scanned reproductions are usually limited in the number of scanning 
lines possible by much the same reasons that a television system is 
limited in the number of scanning lines unless elaborate apparatus is 
specially constructed. Other forms of reproductions have been used 
to simulate television picture structures. Such methods of compari- 

298 E. W. ENGSTROM [j. s. M. p. E. 

son are, naturally, limited to inspection of one picture frame and, as 
such, a still image. 

In television we are concerned with moving images and with a suc- 
cession of movements or scenes which have certain continuity. Also, 
the vision is aided by sound accompanying the picture. Because of 
the wide gap between a still picture of certain detail and a television 
reproduction having the same equivalent detail, it is difficult to draw 
any definite information regarding the number of scanning lines de- 
sired for a particular condition from any of the methods of study 
which have been discussed. These methods are helpful in preliminary 
studies, but fall short when an attempt is made to draw general con- 

Motion in a picture directs the observer's interest to the object or 
objects in motion. Under these conditions the eye requires less 
detail than for a still picture, assuming that the detail is sufficient so 
that the purpose of the movements may be understood. Proper use 
of this may be made in television in the choice of "story action" and 
choice of background for the action. Also, in an image which is the 
result of scanning at the pick-up end, motion of the objects being 
scanned positions these objects for particular frames in favorable 
relation to be analyzed and reproduced when these objects are small 
and approach in at least one dimension the size of the scanning beam. 

For a more complete study of television image, it seems necessary 
to have available the ability to produce image reproductions which 
have picture structures equivalent to television, controllable illumina- 
tion, controllable size, flicker frequency equivalent to television, and 
capacities for subjects which will be used in television. It is also de- 
sirable to cover a range of picture detail equivalent to television im- 
ages of 60, 120, 180, 240, and even larger numbers of scanning lines. 
These equivalents should be so made that they represent nearly per- 
fect picture structures for the detail included. This seems desirable 
so as to avoid mistakes in judgment. Also, it will permit study with 
images equivalent to the more advanced stages of television which will 
later be attained as a result of continued development. Such an ex- 
perimental set-up will allow reasonable determination of several re- 
lated picture properties picture detail, picture size, and viewing 

As has been pointed out, it is impracticable to make use of television 
systems for this study. This is because of limitations in our ability 
at present to produce television images with sufficient detail, illumina- 


tion, and size for this investigation and to have these characteristics 
variable. We must, therefore, resort to suitable equivalents. A 
motion picture film having a picture structure equivalent to a tele- 
vision image provides a very flexible means for carrying out this work. 
Such a method was chosen, and the procedure used will be described. 
There are numerous ways in which such a film may be made, but the 
method used for this investigation is flexible and presents only a 
reasonable amount of preparatory work. 

In the system of television that we are considering, the scanning 
paths are horizontal and the beam progresses from left to right 
(when facing the object or reproduction) and from top to bottom. 
The scanning beam is usually round or square in cross-section. Since 
the scanning beam has width in the direction of the scanning path, a 
certain form of distortion is introduced. This is known as aperture 
distortion, and has been adequately treated in the general television 
literature. This much has been indicated about the image char- 
acteristics because we shall later make comparisons between the 
structure of a television image and the motion picture equivalents we 
are to use. 

The equipment used in making 16-mm. motion pictures with de- 
tail structure equivalent to television images consisted essentially of a 
35-mm. to 16-mm. optical reduction printer. A system of optics was 
interposed between the two picture gates for the purpose of breaking 
up the picture image into small areas, each of which was uniformly 
illuminated, and which transmitted the same total quantity of light 
as a corresponding area in the picture image. A diagram of the opti- 
cal system is shown in Fig. 5. The filament of an incandescent lamp 1 
is focused by means of condenser lenses 2 upon a corrected lens 4. 
Lens 4 in turn forms an image of the 35-mm. picture aperture 3 on the 
plane surface of condenser lens 7. The equivalent of thousands of 
tiny spherical lenses 6 are placed directly in front of lens 7. Each of 
the tiny lenses forms an image of aperture 5. The plane containing 
the many images of aperture 5 is brought to focus upon the 16-mm. 
aperture 9 by means of a corrected lens 8. Condenser lens 7 makes it 
possible for lens 8 to collect an equal quantity of light from each of 
the images formed by lenses 6. The horizontal dimension of the 
rectangular aperture 5 is such that the sides of the images formed by 
lenses 6 just touch, thereby forming continuous bands of light in the 
horizontal direction. The dimension of aperture 5 in the vertical 
direction is narrower, thereby producing narrow dark spaces be- 



[J. S. M. p. E. 

tween the horizontal lines formed. This was done to simulate televi- 
sion image lines. The image at aperture 9 of a motion picture film at 
aperture 3 is broken up by this optical system into as many ele- 
mentary areas as there are lenses or equivalent lenses in 6, each of 
which contains no detail within itself. By adjusting the reduction 
ratios of lenses 4 and 8, and by having sufficient equivalent lenses at 6, 
it is possible to vary the number of picture elements. 

Since it would have been quite difficult actually to obtain the thou- 
sands of minute spherical lenses, an approximate but more practical 

FIG. 5. Diagram of the optical system used in making the motion pictures. 

scheme was resorted to. It is known that two crossed cylindrical 
lenses are very nearly equivalent to a single spherical lens. Thus, it 
would be quite possible to approximate the required condition by 
crossing two layers of fine glass rods, the rods being in actual contact 
with each other. Fortunately, an even simpler solution was found. 
Kodacolor film is embossed with minute cylindrical lenses having 
focal lengths of about 6 mils. By crossing two pieces of Kodacolor 
film with the embossed surfaces in contact, very satisfactory results 
were obtained. The focal lengths of the equivalent spherical lenses 
formed by crossed Kodacolor film were so short that the size of aper- 


ture 5 would have had to be larger than the diameter of lens 4. This 
condition was corrected by forming a cell made up of two pieces of 
Kodacolor film crossed, and filling the space between the embossings 
with a transparent solution having an index of refraction greater than 
air and less than the index of the film base. By varying the index of 
refraction of this transparent solution, it is possible to make the lenses 
have any desired focal length from 6 mils to infinity. 

The Kodacolor cell and lenses 4 and 8 were arranged in a suitable 
mounting and mounted on the reduction printer between the 35-mm. 
aperture and the 16-mm. aperture. Arrangements were provided for 
adjustment of these various lenses. The subject matter was taken 
from a 35-mm. positive print. The first printing operation gave a 
16-mm. negative having the desired picture structure. A 16-mm. 
positive was then made by printing from the negative in a 16-mm. 
contact printer. The sound was transferred in the usual manner. 

Films were made up for a variety of scenes and subjects. These, 
in general, included: 

Head and shoulders of girls modeling hats, 

Close-up, medium, and distant shots of a baseball game, 

Medium and semiclose-up shots of a scene in a zoo, 

Medium and distant shots of a football game, 

Animated cartoons, 


These were assembled for one group with all scenes of the same detail 
(line structure) on the same run of film. For another group these 
were assembled with each scene progressing from 60- to 240-line 
structure. The pictures made included : 

60-line structure, 
120-line structure, 
180-line structure, 
240-line structure, 
Normal projection print. 

It was planned at the start to produce pictures having detail struc- 
tures greater than 240 lines, but it was found that limitations, mainly 
in film resolution, prevented this. The resolution of the 16-mm. 
film used was naturally considerably greater than a 360-line struc- 
ture, but, with the averaging process used in producing each small 
section of the picture, the resolution was not sufficient to pre- 
vent merging of one section into the next. Later determinations 
made from viewing these films indicated that the 240-line structure 

302 E. W. ENGSTROM [j. s. M. P. E. 

pictures were sufficient for the purposes of the investigation since the 
results were of such a nature that the relationship could be extended 
to higher numbers of scanning lines. 

Samples of three picture frames are given as Figs. 6, 7, and 8. 
These are all enlargements from the 16-mm. negatives and include 
structures of 60, 120, 180, and 240 lines and, also, a normal photo- 
graphic enlargement. It is interesting to note how near the 240- 
line structure approaches the normal enlargement in picture quality. 

An RCA Photophone 16-mm. sound projector equipment was used 
in projecting these films. The light cutter in the projector was modi- 
fied so as to interrupt the light only during the time that the film was 
being moved from one frame to the next by the intermittent move- 
ment. This modification consisted in removing one blade from the 
light cutter. The light was, therefore, cut off once per frame, giving 
for these tests a flicker frequency of 24 per second. The films were 
shown to several groups of people, using projected picture sizes 6, 12, 
and 24 inches high. The major reaction from these showings was the 
expression of satisfaction obtained from viewing pictures 12 inches 
high and larger in comparison to smaller pictures. 

It will be of interest at this point to record some of the reactions on 
how well these films form equivalents of television images. These 
reactions were formed as a result of observations and tests made with 
the films. The horizontal-line structure was so clearly equivalent 
that we may pass by this without comment. The changes of contrast 
along the horizontal "scanning" lines for the 60-line structures 
appeared somewhat "mosaic" in arrangement. This was because the 
boundaries of the individual picture arrangements were determined 
by the multiple lens arrangement used to produce the image. This 
effect was not noticed in 120-line structure or in those of higher detail. 
The 120-, 180-, and 240-line structures, and also the 60-line structure, 
except for the effect explained above, were well suited for study of 
image detail. In general, a particular line structure on the film was 
considerably better than a television image (as we are at present able 
to produce them) of the same number of scanning lines. This is a 
desirable condition because the results of the tests will then be in 
terms of television of an advanced stage rather than in terms of 
present capabilities. 

In order to obtain some quantitative information, a number of 
practical viewing tests were made. These tests were made by the 
same three observers who made the tests covered earlier in this paper. 



60 Scanning Lines 

180 Scanning Lines 

120 Scanning Lines 

240 Scanning Lines 


FIG. 6. Sample of picture frame. 



[J. S. M. P. E. 

60 Scanning Lines 

120 Scanning Lines 

180 Scanning Lines 

240 Scanning Lines 


FIG. 7. Sample of picture frame. 



60 Scanning Lines 

120 Scanning Lines 

180 Scanning Lines 

240 Scanning Lines 


FIG. 8. Sample of picture frame. 



[J. S. M. p. E. 

For these observations the same projector equipment as described in a 
paragraph above was used (one light interruption per frame). The 
projection lamp was operated at rated voltage (normal brilliancy) and 
the projection lens was stopped down to give the desired screen il- 
lumination. A screen illumination of five to six foot-candles "was 
chosen. This was measured at the screen, looking toward the pro- 
jection lens, and with the projector running, but without film in the 
picture aperture. This value of screen illumination, though less than 
for theater or home movies, was chosen because it gives a fairly bright 
picture and because it falls within a range to be reasonably expected 
for television. For the pictures of various sizes the foot-candles of 

illumination (surface density) 
was kept the same, varying 
the total luminous flux in pro- 
portion. The stray room illu- 
mination was of the general 
order of one-tenth foot-candle. 
Viewing tests were made 
with projected pictures of 
various heights, using the film 

subjects listed earlier. For 
pictures of given height and 
line structure, observations 
were made for each type of 
subject matter on the film. 
These data were averaged, 

and the information used in the curves to be plotted includes this in 
terms of an over-all average for the three observers. In taking the 
observations, viewing distances were chosen at which the lines and 
detail structure became noticeable. At closer viewing distances the 
picture structure became increasingly objectionable. At the viewing 
distances chosen the picture detail was just satisfactory. At greater 
viewing distances the picture detail was, naturally, sufficient. It was 
noted that the type of picture subject did not influence the viewing 
distance chosen by more than ten per cent. This is explainable on 
the basis that we are determining minimum conditions in terms of 
line and detail structure. Data were taken for pictures 6, 12, and 
24 inches high, and for picture structures of 60, 120, 180, and 240 lines, 
and also for a normal projection print. This information is given in 
curve form in Fig. 9. 





o 100 


A - 6 

B - 12" 
C - 24" 













h ^"***. 

- . 










* 4 6 8 10 IS 14 16 IB 80 2 24 

VIBRIO Dxnaci it FZBT 

FIG. 9. Viewing tests with pictures of 
various heights. 

May, 1934] 



In order to present these observed data in more general form, the 
above curves are shown, in Fig. 10, replotted in terms of scanning lines 
per inch. The curve drawn through the observed points is the two 
minutes of arc curve from Fig. 2. This, as explained earlier in the 
paper, is a curve between scanning lines per inch and viewing dis- 
tance where the dimension between centers of scanning lines subtends 
an angle to the eye of two minutes. Because of the correspondence 
between the plotted points and the two minutes of arc curve, we shall 
use this two minutes of arc curve in our discussion as representing the 
average results (for the three observers) for practical viewing condi- 

It is of interest to compare these observed results from viewing the 
films of several detail structures with the observed results of viewing 
the chart, Fig. 1, the curves of 
which are shown in Fig. 4. 
In the case of the still chart 
observations, the average falls 
on the one minute of arc 
curve; in the case of the ob- 
served motion picture tele- 
vision equivalents, the aver- 
age falls on the two minutes 
of arc curve. 

In order to indicate the 
relative viewing distances for 
a normal projection print, the 
data taken are shown in an- 
other graphical form. The plotted points for picture structures of 60, 
120, 180, and 240 lines are the same as for the curves in Fig. 9. The 
plotted points for a normal projection print were taken in the same 
manner as for the other films. In this instance the viewing distance 
chosen was where the picture just began to show loss of detail. Thus 
Fig. 11 indicates in a general measure the relative merits of the several 
picture structures. It also shows how near the 240-line structure ap- 
proaches a normal 16-mm. projection print. In inspecting this chart 
it will be noted that the observed data do not entirely check the theo- 
retical acuity conditions. This is also to be noted by the variation of 
the points from the theoretical curve in Fig. 10. An example of this 
for the chart, Fig. 11, is that curve D for 240 lines should indicate 
viewing distances one-half that for curve B of 120 lines. Curve B 

C 4 6 8 10 12 14 16 18 20 2 24 

FIG. 10. Data of Fig. 9 replotted. 



[J. S. M. p. E. 

for 120 lines and curve A for 60 lines show the proper one-to-two re- 
lationship for viewing distances. It is probable that for the higher 
number of lines the observed data err on the side of being too "good." 
With a screen illumination of the order used in these tests (5 to 6 
foot-candles) an increase in apparent detail can be obtained with 
higher values of illumination, thereby providing a greater range of 
contrast. To determine the general order of this increase, several 
tests were made with a screen illumination of 20 foot-candles. With 
this value the apparent picture detail was improved, but also the 
picture structure was more pronounced, requiring a choice of viewing 
distance from thirty to forty per cent greater than for an illumination 

of 5 to 6 foot-candles. Since 
5 to 6 foot-candles are more in 
keeping with television possi- 
bilities for the next several 
years, and since the difference 
in apparent detail and view- 
ing distance is within the 
| // /' .*'' ^ accuracy tolerances of the 

- L ' '-' generalizations to be drawn, 

this particular condition will 
not be taken into account. 

Some interesting data were 
obtained from direct compari- 
sons of these projected tele- 
vision equivalents with the 
same subjects having "per- 
fect detail." Two similar 

projectors were set up so that the projected images were side by side 
on the screen and the illumination of each the same. One projector 
was used to project the film having television line structure; and 
the other projector, the same film subjects but a normal projection 
print. Observations were made using pictures of several heights and 
using the films having picture structures of 120, 180, and 240 lines. 
Viewing distances were chosen at which the two screen images had 
the same apparent detail. At these viewing distances the image 
from the normal projection print had a detail structure beyond eye 
acuity in fineness, and in this sense "perfect." These viewing dis- 
tances might, therefore, be termed "ideal viewing distances" from 
the standpoint of picture detail and structure for the television 

t 4 6 8 10 12 14 16 18 80 84 

FIG. 11. Chart showing the relative 
merits of the several picture structures: 
A, 60 scanning lines; B, 120 scanning lines; 
C, 180 scanning lines; D, 240 scanning lines; 
E, projection print. 

May, 1934] 



equivalents. The data taken indicate that this "ideal viewing 
distance" is approximately fifty per cent greater than the "minimum 
viewing distances" shown by the curves in Figs. 9 and 10. This in- 
formation is shown in graphical form by Fig. 12. 

We have determined from these observations two viewing distances 
in terms of picture detail and structure. The first is a minimum view- 
ing distance, and the second an ideal viewing distance. If the total 
picture size were limited, we would, in viewing this picture, tend to 
approach it until the picture detail and structure became unsatis- 
factory. We would for this condition choose the minimum viewing 
distance referred to above. 
If the total picture size were 
ample, we would tend to posi- 
tion ourselves so that we 
would view it at the ideal 
viewing distance. This re- 
lationship will be covered 
more fully later in the paper. 

The tests we have made on 
picture detail are rigorous. 
We have set as standards the 
ability of the eye to see the 
elements of detail and picture 
structure. Another less ex- 
acting standard would be 

the "ability ' of images having various degrees of detail to "tell the 
desired story. ' ' In this case the detail required is dependent upon the 
kind of story and information to be presented. The detail require- 
ments would increase as the scenes became more intricate. During 
the early stages of development such a standard is useful, but, for 
obvious reasons, it is not of a lasting type since it is the eye and the re- 
actions of vision that must be satisfied. The standards we have used 
are definite and of a character which will not become obsolete as the 
development of television progresses. 

If we qualify and limit "the ability to tell a desired story" to specific 
conditions, the experience we have had with television and these tests 
allows us to make some interesting approximate generalizations. If 
we take as a standard the information and entertainment capabilities 
of 16-mm. home movie film and equipment, we may estimate the 
television images in comparison. 

4 6 8 10 18 14 16 16 SO 22 24 


FIG. 12. Comparison of projected tele- 
vision equivalent with same subjects having 
"perfect detail." 

310 E. W. ENGSTROM [j. s. M. P. E. 

60 scanning lines entirely inadequate 

120 scanning lines hardly passable 

180 scanning lines minimum acceptable 

240 scanning lines satisfactory 

360 scanning lines excellent 

480 scanning lines equivalent for practical conditions 

This comparison assumes advanced stages of development for each 
of the line structures. It relates to the ability of observers to under- 
stand and follow the action and story. It does not relate to the 
ability to reproduce titles and small objects. 

We stated earlier in this paper that motion in a picture has an effect 
on the apparent detail. There are several reasons for this. The 
observer's interest is directed to the object or objects in motion. The 
eye then does not tend to explore the picture step by step, examining 
each section critically. Under these conditions the eye requires 
less detail than for a still picture, assuming that the detail is sufficient 
so that the purpose of the movements may be understood. Objects 
made up of too few picture elements to recognize while still, may be 
recognizable and realistic while in motion. A portion of this im- 
provement is due to experience on the part of the observer in associat- 
ing the motion with things and processes he understands. A portion 
of the improvement is due to more favorable conditions for scanning 
while the object is in motion. Another portion of the improvement, 
as already stated, is due to concentration of interest around the 
motion. This effect is very important in dealing with crude television 
images, but becomes minor in images having sufficient detail to satisfy 
eye acuity. An image made up of 30 scanning lines, though inade- 
quate for almost any subject, provides much more satisfactory results 
for objects in motion than for still scenes. On the other extreme, a 
normal 16-mm. projected image of a scene including motion is not, 
in any large measure, superior to a scene containing no motion. 
There is, of course, a decided difference in the center or centers of 

Reference to Figs. 6, 7, and 8 will illustrate this. In particular, in 
the 60-line print of the baseball scene, the players are about five pic- 
ture elements high, and considerable imagination must be used to lo- 
cate them. With the same scene in motion the observers can pick out 
the players, roughly determine their action, and, in a general sense, 
follow the game. In other words, the condition has changed from a 
reproduction of a scene containing no motion, and which gives prac- 


tically no information except that it is a baseball field, to a reproduc- 
tion of the same scene in which the players move, and which in general 
allows the observers to follow the action roughly. It is apparent from 
examining the other prints, particularly as the amount of detail in- 
creases, that reproductions with motion would naturally improve the 
satisfaction obtained, but the difference would not be as great and 
would decrease as the picture detail improved. Summarizing the 
effects of motion in a television image, we may conclude that the 
major improvement is that of observer interest. This is true because, 
to be generally satisfactory, the image must contain sufficient detail 
to satisfy eye acuity. This same condition holds in the case of motion 
pictures. We are, therefore, justified (and safe, from the standpoint 
of results) in discounting the effects of motion in the generalizations 
to be drawn from this analysis. 

Thus far in our investigation we have considered picture detail and 
structure and have arrived at certain relationships between number 
of scanning lines and viewing distances. We have not taken into 
consideration the picture size. By reference to the curve in Fig. 10, 
and by knowing the total number of scanning lines available for the 
system we are considering, we may readily determine the size of 
the picture in terms of height. This does not, however, tell us, at the 
viewing distance we have chosen, that the picture will be of a size 
pleasant to view. If the picture is too small it will be unsatisfactory 
because too fixed an attention will be required for viewing. If the 
picture is too large it will be unsatisfactory because too large move- 
ments of eyes or head will be required for viewing. In television, be- 
cause of the practical limitations in detail (scanning lines), we are 
confronted in general with too small rather than too large pictures. 

In television we use the same ratios of picture width to picture 
height (aspect ratio) as in motion pictures (6 to 5, or 4 to 3) . In moder- 
ately large theaters the distance from the back row of the orchestra 
section to the screen does not usually exceed six to seven times the 
screen height. The front row of seats may be as close as one and one- 
half to two times the screen height. The choice position is probably 
at four times the screen height from the screen. In home movies 
(where less detail is available because of the smaller size film) the 
desired viewing distances cover a span of from four to eight times the 
picture height. Since television, of the type we are considering, is 
for home entertainment, we shall in this consideration of television 
picture size use the accepted ratio of picture height to viewing 



[J. S. M. p. E. 

distance for home movies (span of one to four one to eight) in our 
comparisons. To make this more specific we shall follow through an 
example. For this illustration we shall use a picture one foot high. 
The desired viewing range is from four to eight feet. Going beyond 
eight feet, viewing conditions become decreasingly satisfactory and 
at twelve feet and beyond become quite unsatisfactory. This is 
based on the assumption that the same general run of subject matter 
will be used as for motion pictures. 

We have now accumulated data which allow preparing a chart in- 
cluding relationships between scanning lines, picture size, viewing dis- 
tance, and desired ratios 
/ 4 of picture height to view- 
ing distance. The infor- 
mation on this chart, 
which is given as Fig. 13, 
is based on the observed 
data recorded in curve 

*t- <*?. .* -u. form in Fig. 10. Using 
I J 1 llnlfUf^J^rG^ 20 this "minimum viewing 

distance" relation be- 
tween scanning lines per 
inch and viewing dis- 
tance, the chart in Fig. 
13 shows for a number 
of viewing distances the 
picture size total scan- 
ning line relationship. 

Superimposed on this are horizontal broken lines for picture height to 
viewing distance ratios of one to four, one to eight, one to twelve, and 
one to sixteen. In using this chart we must take into consideration 
the fact that between the one-to-four and one-to-eight picture height 
to viewing distance lines, the viewing conditions will be satisfac- 
tory. As we drop below the one-to-eight ratio line the viewing 
conditions become less satisfactory, and below the one-to-twelve 
ratio line, generally unsatisfactory. 

This chart (Fig. 13) includes all the necessary information to de- 
termine scanning lines required if viewing distance and picture height 
have been decided upon ; or picture size, if a certain number of scan- 
ning lines are possible and a certain viewing distance is desired. The 
chart also provides a guide for the desired picture sizes for general 


FIG. 13. Relationship between scanning lines 
and picture size for several viewing distances. 
Broken lines indicate picture height to viewing 
distance ratios. 


viewing conditions. To illustrate, we might decide that we wish 
to view a television image at eight feet. Starting down the eight-foot 
viewing distance line, we find that with 360 scanning lines we may 
have a picture twenty inches high. We also learn that the picture 
height to viewing distance ratio is a very desirable one. With 240 
scanning lines we find that we may have a picture thirteen and one- 
half inches high. Here the picture height to viewing distance is 
above the one-to-eight ratio line and, therefore, satisfactory. With 
180 scanning lines we may have a picture ten inches high. We note 
that we have dropped below the one-to-eight ratio line, a less desir- 
able viewing condition. At this point the picture will, in general, 
be satisfactory for viewing but probably the minimum desirable for 
an eight-foot viewing distance. 

The viewing distance lines on this chart mean that, at this par- 
ticular distance and for the number of scanning lines and picture 
height indicated at any point along the line, this is the minimum view- 
ing distance for a picture of this number of scanning lines and height. 
Since this information is based on tests made by three observers who 
have, as previously pointed out, acuity above average, this is a safe 
condition for average use. Suppose, as in the above illustration, we 
have chosen an eight-foot viewing distance and, with 240 scanning 
lines available, a picture height of thirteen and one-half inches. The 
nearest an observer should view this image is, then, at eight feet. 
To determine if the general viewing conditions at more distant 
points are satisfactory because of the picture size, we may start 
at the eight-foot viewing distance line and the thirteen and one-half 
inch picture size and drop down along the thirteen and one-half inch 
ordinate. At a ten-foot viewing distance we are just a little under 
the one-to-eight ratio. At a twelve-foot viewing distance we are 
nearing the one-to-twelve ratio and approaching unsatisfactory 
viewing conditions. Therefore, a picture of 240 scanning lines and 
thirteen and one-half inches high may be viewed from eight feet to 
about twelve feet. 


The author expresses appreciation for the assistance of Messrs. W. 
L. Carlson and T. V. DeHaven who together with the author were 
the three observers referred to in the paper, for the assistance of 
Mr. T. V. DeHaven in preparing some of the test set-ups, and for the 
assistance of Mr. G. L. Dimmick in preparing the films having the 
special detail structures. 



Summary. On January 30, 1934, a demonstration of the transmission and repro- 
duction of speech and music in auditory perspective was held in the auditorium of the 
Engineering Societies Building, New York, N. Y., to which the members of the 
Society of Motion Picture Engineers and the Acoustical Society of America were 
invited. The following are Dr. Fletcher's remarks in explanation of the various 
features of the demonstration. 

We have met tonight to listen to some demonstrations of sound in 
"auditory perspective." There is nothing mysterious about the 
word, although it might seem like mixing sight and sound. Per- 
haps even that is permissible, particularly before a motion picture 
engineer's society. What we mean when we use that term is a 
system that will project sound into an auditorium in such a way that 
you are able to recognize the direction from which the sound is com- 
ing, that is, to be able to locate the sources of sound on the stage. 
Probably as you have already noticed, if you listened critically to 
the piece just rendered, the music was coming from all parts of the 
stage rather than from a single point, as would be the case if a single 
loud speaker were used for reproducing. This characteristic of 
auditory perspective is not the only one that we think is important 
in this system. No new principles have been suddenly discovered 
and incorporated into this system, for the principles necessary for 
building such a system were worked out some time ago. It was only 
after considerable research, however, that we were able to embody 
them into concrete physical apparatus to the degree that was desired, 
so as to fulfill all the requirements for reproducing any sort of sound, 
though it might contain frequencies from the lowest to the highest 
that are audible, and intensities from the lowest that you can hear 

* Presented at the Engineering Societies Building, New York, N. Y., January 
30, 1934. 

** Bell Telephone Laboratories, New York, N. Y. 


to the highest that you can tolerate. This sytem was originally 
designed for reproducing symphonic music. 

The requirements for reproducing any sort of sound, you might 
think, would depend upon the character of the sound. That is true 
to a certain extent. For example, if you wanted only to reproduce 
a sound that contained no components except within a limited range 
of frequency, say, from 500 to 2000 vibrations per second, it is evident 
that the system would be required to reproduce only that range. 
However, if you wished to reproduce even such common sounds as 
the clapping of the hands, or any sound that is started suddenly or 
stopped suddenly, we know from theoretical consideration that it 
would be necessary to reproduce an infinite range of frequency to re- 
produce it correctly. Consequently, impact sounds such as the 
clinking of keys or the ringing of bells, theoretically, require an in- 
finite range of frequency. Inasmuch as such sounds form a large 
class the requirement imposed upon the system is not determined 
by the source of the sound but by the limits of hearing. If there are 
components in the sound that we can not hear, it is not necessary 
to reproduce them. If we desire to design a system that will repro- 
duce any sort of sound that we can hear, then we must design the 
system so that it will be able to reproduce all frequencies from 20 to 
20,000 cycles per second, inasmuch as that is the range of hearing. 

For symphonic music, however, it is not necessary to reproduce 
such a wide range. We determined by actual hearing tests, that is, 
by listening to an orchestra with a very wide-range system and then 
limiting the range by means of filters, that if we eliminate fre- 
quencies beyond 15,000 cycles per second, or below 40 cycles per 
second, there is no detectable difference in the reproduced music. 
The system that is being demonstrated tonight was designed from that 
viewpoint, for reproducing symphonic music, and consequently re- 
produces faithfully all frequencies from 40 cycles per second up to 
15,000 cycles per second. For the same reasons that the hearing 
limits the range of pitch, it also limits the range of intensity. If 
we reproduce the orchestra merely as it is played, it will require a 
range of about 60 to 70 decibels. This range does not use the entire 
audible range. 

Measurements of hearing indicate that, when the intensity level 
reaches 120 decibles, the "feeling point" is reached, for most of the 
frequency range, corresponding to an intensity of 10 ~ 4 watts per 
square centimeter of the wave-front. Although the entire upper 



[J. S. M. P. E. 

part of that range can be used the lower part is limited, because 
there is always some noise in the hall where the sound is reproduced. 
For example, if I asked you to remain as quiet as humanly possible, 
there would still be enough noise, due to breathing and shifting of 
clothing, etc., so that you would not be able to hear a sound of as 
low an intensity as you would if you were shut up in a sound-proof 
room where there were no such noises. So it is difficult to use the first 
20 decibels above the threshold. 

1 - 






/ / 

^ O 

Zp i / 


x / x 



. v / / 

r / / 

/ / /o 

/ / 



2048 o 



256 > 



FIG. 1. 

Chart showing relation between frequency, octaves above 
and below middle C, and musical notation. 

A range of approximately 100 decibles, or a range of 10 l in in- 
tensity ratio, is the useful range of hearing. That is a very large range 
compared with that of a good many reproducing systems in use to- 
day. Phonographs and radio sets have a range of something like 
10 4 . 

The first experiments to be performed will show you what we mean 
by auditory perspective, and how a sound source may be apparently 
projected on to the stage and moved about. We shall then demon- 
strate the second characteristic, namely, the ability to reproduce sound 
through the entire audible pitch range from the lowest that you can 


hear to the highest that you can hear. Next, we shall perform some 
other experiments that will show that this system can utilize the 
entire range of useful hearing from sounds as soft as you can 
hear to sounds as loud as you can tolerate. Finally, we shall demon- 
strate some effects achieved by using the electrical controls during 
the reproduction of the sound. 

As most of you know, our transmission system is composed of 
microphones to pick up the sound, of amplifiers, and of loud speakers 
or sound projectors to reproduce it. The microphones are located 
on the fifth floor of this building where we have a miniature stage on 
which the action takes place, and also another small stage on which 
an orchestra of 25 pieces will play. By means of wires the micro- 
phones are connected to loud speakers on the stage before us; and 
all the sound that you hear coming from behind the curtains is repro- 
duced electrically. Let me make that statement emphatic, because, 
in spite of repeating it over and over again, we have found by experi- 
ence that people go away from our demonstration thinking that some- 
one was behind the curtain. 

In the first demonstration we have an act that illustrates the 
property, auditory perspective. Certain noises and sounds will be 
produced on the miniature stage, the main object of the demon- 
stration being to enable you to determine the position of their 
sources. There is also the other object of showing you that the sys- 
tem will actually reproduce faithfully sounds that you do not hear 
reproduced faithfully by any other transmission system. They are 
the impulsive sounds that I mentioned; I am sure you will be able to 
recognize them as the experiment progresses. 

[Then followed the reproduction of a man hammering a nail into 
a board, and sawing wood; the sources of the sounds assumed various 
positions on the unseen stage, and the reproduction of the sound by the 
system produced an auditory effect corresponding to the positions of the 
sources Ed. ] 

In the next demonstration we shall have two trumpeters : one here 
as you see before us, and one on the miniature stage. One of the 
trumpeters will play here, and then the one upstairs, or vice versa; 
and then they will move about the stage. The trumpeter here, you 
will see move; but try to determine where the other man is moving 
and determine, if you can, which one is playing. 

[The demonstration described was very effectively performed Ed.] 

That demonstrates what we mean when we say "auditory per- 

318 HARVEY FLETCHER [J. S. M. p. E. 

spective;" you had no difficulty in locating the position of the trum- 
peter on the miniature stage on the fifth floor. 

As I said before, the sounds most difficult to reproduce naturally 
are the impact sounds, such as the sound of a bell or any sound that 
is caused by percussion. In the next demonstration we shall use 
two such sounds, one from the tambourine, which, as you know, is 

FIG. 2. One of the reproducer units used in the demonstration. 

produced by striking a stretched skin and by the rattling of the little 
clappers on the side, and one from the tinkle of a bell. Inasmuch 
as they require such a large range of pitch to make them sound natural 
they are very directional, because high frequencies are much more 
directional, than the low frequencies, so I hope in this demonstration 
you will be able to follow without difficulty the locations of the 
sources. [Demonstration] 


I neglected to say that you get the sense not only of right and left, 
but also of forward and backward. Of course, when you are a 
great distance from the stage the forward and backward motions are 
difficult to detect, even when the sound sources are on an actual 
stage before you. Those in the front part of the house will have no 
difficulty in locating the sound as back stage or front, as well as to 
the right and to the left. 

In the next demonstration we shall again use an impact sound. 
It is what we call the illusion of a tap dancer. One of the tap dancers 
will perform here and one upstairs. See whether you can tell by 
the sound which one is dancing. [Demonstration] 

There are many possibilities of producing rather peculiar and 
amusing effects by operating the electrical controls. For example, 
by sliding the dial backward and forward, you can emphasize the 
rhythm of the tap and also make it any loudness you desire, as we 
shall now demonstrate. [Demonstration] 

Another possibility of the system is that of reproducing the voice 
of a singer at various positions on the stage. It therefore opens some 
possibilities for opera. If desired, one might have some pantomime 
actors on a stage going through the action, and the singers off the 
stage producing the music. In this particular demonstration the 
singer is moving about on the miniature stage in the corresponding 
positions which you will recognize here on this stage. [Demonstration ] 

Those of you who were at the meeting of the Acoustical Society 
a little over a year ago, remember that Dr. Stokowski gave a talk 
in which he said that in the future he expected to see opera put on in 
pantomime and that he would have the good singers back stage and 
the good looking performers on the stage doing the pantomime. This 
demonstration, I think, shows the possibility of doing so if desirable. 

In the next demonstration we wish to show the possibilities for 
drama, as well as for opera. I think that there are a number of pos- 
sibilities for the system in legitimate drama or in sound pictures. 
Just how it might be worked out I do not know, but in the next demon- 
stration we shall present a brief dramatic sketch in which the actors 
speak from various positions on the distant stage. There is some 
quick action at the end of the act but I want you to notice particularly 
the fidelity of the pistol shot that ends the act. 

[Demonstration: The voices emanating from the loud speakers ap- 
parently followed the actors about in their movements on the unseen 
stage Ed. ] 

320 HARVEY FLETCHER [j. s. M. p. E. 

Those of you who have had some experience in trying to repro- 
duce the sound of a pistol by electrical transmission know that it 
is very difficult to do, and I assure you that what you heard is a rather 
faithful reproduction. It is against the law to use firearms in which 
an actual bullet is used. This gun uses a blank cartridge. We had 
it down here on the stage and it sounded like what you heard. In 
the next demonstration also we shall have a pistol shot, but one of 
our own make rather than with a real pistol. We call it a pistol 

FIG. 3. The control station in the balcony of the auditorium. 

shot, but it is something to illustrate how we can make the sound 
travel across the stage. You will hear something that sounds like 
a report of a gun, the bullet will come out leisurely, go across the 
stage and strike the target. This will be repeated, then the bullet 
will come out of the target and return to the gun. 

[Demonstration: The sound of the shot came from one side of the 
stage, and that of the shot striking the target from the other. A whistling 
sound seemed to follow the "bullet" across the stage Ed.] 

I might explain how the act was done. A man on one side of the 
miniature stage on the fifth floor clapped two boards together, an- 


other ran across the stage, blowing a whistle; the man on the other 
side then struck a metal circular saw that was handy; and you heard 
the result. We used that particular stunt, for it is just a stunt, be- 
cause it shows rather vividly what we mean by auditory perspective. 

Now if we have only a single source of sound on the stage, that is, 
a single source at a time, it is very easy to produce the illusion of 
motion without having any motion on the miniature stage at all, 
by simply manipulating the electrical controls. The apparent posi- 
tion of a source of sound on the stage is dependent upon the relative 
loudness of the sound that reaches your ears. When more sound 
comes into your right ear, you localize the sound as on the right. 
When more sound comes into your left ear, you place it on the left; 
for that reason we can easily manipulate the dials controlling the 
amplifiers connected to the loud speakers, so as to make the sound 
move either right or left. The possibility of making it move back- 
ward, however, is a little more difficult, for in that case it is necessary 
not only to decrease the sound intensity but also to add reverbera- 
tion to it. In the demonstration only transverse motion took place. 
We shall now put on the same demonstration by manipulating the 
controls. Remember this time that the sounds, clapping of the 
boards, the blowing of the whistle, and the striking of the circular 
saw, all occur before a single microphone, and all the apparent 
motion is produced by manipulating the dials. [Demonstration] 

It is evident that the apparent location of the voice of a singer can 
also be controlled by means of the dials. Our artist will next sing 
the same song she sang a few minutes ago, but this time will remain 
stationary in front of the microphone; the apparent movement will 
be produced by manipulating the dials on the control. 

[Demonstration: The voice of the singer was made apparently to move 
from one point to another on the stage Ed.] 

In these demonstrations we have emphasized the property that 
we call "auditory perspective," but, of course, we have used a sys- 
tem that has other characteristics that must be practically perfect 
or you would not realize the effects that you have heard. In the 
next demonstration, we shall use the full orchestra. We shall use 
the wide band system, reproducing all the frequencies in the orchestra 
uniformly, and we shall use the controls so as to emphasize the con- 
trasts in the music and do it as artistically as we know how. A 
musician will manipulate the dials, so that the result that you will 
hear will be due to the composer first, then to the players in the or- 

322 HARVEY FLETCHER [j. s. M. P. E. 

chestra, then to the director's interpretation, and finally to the in- 
terpretation of the person manipulating the dials. The selection will 
be Shubert's Unfinished Symphony. In the first part the music 
comes from the full stage, and you will be able to localize the instru- 
ments if you try. [Demonstration] 

You will all agree that for an orchestra of only 25 pieces, with only 
four first violins, the music sounds considerably like that of a large 
orchestra. That demonstrates the possibilities of using the controls. 
Even with a large orchestra such as a symphony orchestra, this sys- 
tem could be used, we think, to improve the music. I think a large 
number of musicians at our Washington demonstration agreed with 
this conclusion although some disagreed. But just what use would 
be made of this stereophonic system in music, must be determined 
by the musicians after they have a chance to use it and to compose 
their music with it in view. 

For the benefit of those who are interested in the engineering aspects 
of this demonstration, the controls are in the balcony. Mr. Snow, 
who is at the controls, can hear everything that is going on in the 
room, and controls the volume and the quality of the sound. The 
loud speakers are behind the screen here on the stage. The ampli- 
fiers are off-stage at the right, and there is a control man located here. 
Then, of course, there are the controls on the fifth floor, where the 
miniature stage is located; and, finally, there is another group of men 
in the projection room for performing some further experiments, 
which you will see later. All these various places are tied together 
by means of telephone lines. I have a microphone here on the 
speaker's stand so that all the groups are hearing what I say, and in 
that way the demonstrations are synchronized. 

Next, let us turn our attention to the second characteristic of the 
system, which we feel is worthy of notice: namely, that it will re- 
produce with almost uniform efficiency frequencies from those so 
low that you can scarcely hear them to those above the audible range. 
As I told you at the beginning, to produce such sound as we have 
been producing here requires a range from 40 to 15,000 cycles. In 
order to convince you that the system will actually reproduce such a 
range, we are going to reproduce single frequencies, or pure tones, 
beginning with the lowest pitch and ascending to the highest audible 
pitch. To aid you in following the experiment, we shall use a 
chart, which you now see on the screen. In the projection room is a 
heterodyne oscillator, which will generate currents of 40 to 15,000 


cycles. The tuning condenser of the oscillator is geared to the 
indicator that you see projected across the chart, the position of the 
latter on the chart indicating the frequency of the oscillator. The 
heterodyne oscillator is connected to the amplifier controlling the 
loud speakers. 

We shall first reproduce the sounds starting at the lower limit of 
40 cycles and then proceed to the highest, 15,000 cycles. At various 
pitches as the frequency rises a number of you will be unable to hear 
the sound. In any audience of this size there will be persons whose 
hearing begins to be defective anywhere from 2000 cycles up to 15,000 
cycles. This is not only a test of the system, but it will also be a 
test of your hearing. [Demonstration] All right, we'll start. 

On the left-hand side of the chart you see a pitch scale, which 
indicates the number of octaves above or below a thousand cycles 
per second. That reference frequency corresponds approximately 
to high C on the musical scale. At the center of the chart is a musical 
scale from which you can read the pitch. The frequency is indicated 
by the right-hand scale. The tones that you just heard were pro- 
duced with practically the same intensity. When listening to the 
low tones you probably thought there were peaks in the system, but 
that was due to the room, the reflections of which build up intensity 
maxima in certain spots, and as you change the pitch those spots 
switch around the room. You probably noticed also that some low 
frequencies caused rattling from somewhere in the room. I assure 
you it was not from the stage. It was from the window panes and 
various panels in the room which were resonant to the frequency we 
were producing. If we make that frequency loud, you will notice 
that it will shake the panels very vigorously. For fear of bringing 
some of them down, we shall not hold the tone at one particular fre- 
quency very long, but I should like you to hear what intense sounds 
can be produced. We shall go so rapidly through the pitch range 
that there will be little danger of any resonance being too strong. If 
you watch the indicator you can see the range of variation. [Demon- 

This would be a good test for locating bad resonances in audi- 
toriums. I am sure that in a good many halls such bad resonances 
exist without their being detected. They probably produce bad 
effects on the music if it is very loud. But I think in the future 
when this reproducing system it to be used with such great intensity 
an exploration for locating such bad resonances must be made, and 


the offending member rectified before you can expect to have a room 
of very good acoustics. 

There are other ways of showing the pitch range of the system. 
In the last experiment, we produced a single tone and made its pitch 
go through the whole range. Another way that might show the 
same effect is to produce a sound that has all the frequencies present, 
such as any of the impact sounds of which I spoke, and then limit 
the range by means of electrical filters. One can introduce into the 
transmission circuit electrical filters that will eliminate all frequencies 
above a certain value, or all below a certain value. In the next two 
experiments we shall do that, using the chart as before. The chart 
has been changed somewhat; at the bottom is the vibration number 
and the corresponding note on the musical scale. For example, 
high C and 1024 correspond. The blackened portion which you see 
there represents the part that is eliminated by the filters in our trans- 
mission system. First you will hear the sound reproduced naturally, 
that is, with the complete range. Then the filter will eliminate the 
frequencies represented by the blackened portion of the chart, and 
you can compare it with the original sound. The first source of 
sound that we shall use will be an impact instrument. It will be the 
snare drum, one of the most difficult instruments to reproduce, 
accompanied by a fife. The players will first march across the stage 
as they play, to show you how naturally such instruments can be 
reproduced. Then they will stand in the center while we perform 
filtering operations on the sound. [Demonstration] 

You can judge from what you have heard what range is necessary 
to reproduce the snare drum and the fife. If your judgments were 
all written down and compared you would find that observers differ 
because the range depends on hearing ability. If you can hear 
frequencies above 8000 cycles you can readily detect the introduction 
of the 8000-cycle low-pass filter; but if you can not hear frequencies 
above 8000 cycles, you can not detect its introduction. 

In the next experiment the full orchestra will be used, and similar 
filtering operations will be performed. In this case, as I told you 
before, for a large orchestra at least, a range from 40 to 15,000 cycles 
is required to make all the instruments sound natural. You can 
judge for yourself whether this orchestra requires the full range or 
not. We shall filter out the top frequencies and then the bottom. 
When the chart on the screen is completely white the system is re- 
producing uniformly from 40 to 15,000 cycles. When part of it is 


blackened, the frequency band corresponding to the blackened por- 
tion of the chart is being filtered out. [Demonstration] 

Experiments of this sort are a little unsatisfactory before a general 
audience, particularly when you have to depend upon someone else 
to throw the filters, because some of you may feel they are changing 
at the wrong time. Everybody has his own notion as to when he 
would like to have the filters switched. It was through a series of 
experiments of just this type, using a jury of 15 or 20 experienced 
listeners, some of them musicians of national repute, listening to the 
Philadelphia Orchestra under the direction of Dr. Stokowski, that 
we were able to come to the conclusion that eliminating frequencies 
below 40 cycles or above 15,000 cycles did not produce a change that 
could be detected. But at 13,000 cycles, for example, there was a 
very definite majority opinion that something had been changed in 
the music. 

In the next series of demonstrations we wish to illustrate the third 
property of which I spoke, namely, the wide intensity range that this 
system is capable of reproducing. I said that to take advantage of 
the entire range of hearing would require a range of 120 decibels 
on the intensity level scale. Now those of you who are familiar with 
the theory of room acoustics know that there is a definite relation- 
ship between the intensity of a source of sound in a room, the volume 
of the room, and the absorption in the room. As you can readily 
see, if the absorption is greater it will require more power to build 
up a certain intensity in the room. Also, if the room is large, it re- 
quires a greater power. Those relationships have been worked out 
mathematically and, if we put into the resulting equations the 
threshold intensity for feeling, then we arrive at a relation that gives 
the intensity of sound that the loud speakers must produce in terms 
of the volume of the hall and its reverberation time. That relation- 
ship was given in one of the papers presented before the American 
Institute of Electrical Engineers. Using that relationship, it turns 
out that for a hall like the Academy of Music in Philadelphia, the 
power that the loud speakers must produce at their peak is about 
half a kilowatt, 400 watts to be more exact. That, then, is the 
maximum power that the loud speakers will ever be required to 
deliver in a hall of that size. As a matter of fact, if there is any 
greater capacity than that there is danger of impairing somebody's 
hearing. The loud speakers we are using will produce about that 
amount of acoustic power. The electrical power put into them by 

326 HARVEY FLETCHER [j. s. M. P. E. 

the amplifiers is, of course, greater than that. The approximate 
efficiency for musical sounds is about 50 per cent. In the loud 
speakers handling the low range it is as great as 75 per cent, so that 
the electrical power required for the loud speakers is approximately 
1 kilowatt for producing this maximum sound intensity. 

The meaning of such an intensity range will be illustrated by a 
few experiments. To follow the experiments we shall use a chart 
that has an intensity scale rather than a pitch scale; the intensity 
level scale is expressed in decibels, from the threshold of hearing to 
the threshold of feeling, which is approximately 120 decibels. Near 
the intensity levels corresponding to the top of the scale you begin 
to feel the sounds and, of course, you can not get much lower than 
20 decibels before you begin to be troubled by the noises in the room. 
The little pointer at the side of the chart, which indicates the level 
that is being produced, is geared directly to a level recorder, so that 
the sound intensity level in the room can be read directly on the 
scale. By means of a microphone just beneath the balcony at the 
rear, the sound is picked up and then put through an amplifier, 
from which it goes to the level recorder, and the indicator moves up 
and down as the intensity of sound changes. It reads the intensity 
level in front of the microphone. That is not the same as in other 
parts of the room, although it will give you a rough indication of the 
intensity of the sound where you are sitting. In any room, no matter 
What kind of sound you produce, due to the reflections from the walls 
there are certain places of maximum and others of minimum sound. 
I notice that you are already trying to assure yourself that the indi- 
cator will respond to sounds that you make as well as to the sounds 
of my voice. This level indicator has a range slightly greater than 
90 decibels. Consequently we must choose the range that we wish to 
cover. Since that is the upper range, its resting point is somewhere 
between 20 and 30 on the scale. Usually there is enough sound in 
the room to keep it up to that level anyway. 

The scale is the one that the Acoustical Society of America has 
adopted recently and the American Standards Association is now con- 
sidering for adoption. It has a zero that corresponds to an in- 
tensity of 10~ 16 watts per square centimeter and a scale in decibels. 
As I am talking now the actual intensity level, if I sustain my voice 
like this: "Ah," should be between 40 and 50. That is about the 
level of the voice of a speaker from the platform into a large room of 
this sort. Remember these levels are the levels produced out in 


front of the microphone. On the chart, ordinary conversational volume 
refers to, of course, the intensity level that would be received by a 
listener two or three feet away from the speaker. Above that is the 
singing intensity level of a loud soprano in a concert hall, which you 
can check when the singer sings. If we desire, by using the elec- 
trical controls we can raise the intensity of the singing voice much 
above that level. 

In the first demonstration, we shall use a sound source to check 
the scale. As you see, aeroplane noise should be about 110 decibels. 
Let us listen to it at about its natural intensity if it came close to us. 
The measurement indicated on the chart was made in the cockpit 
of the aeroplane. Now we shall see if we can reproduce the aero- 
plane noise at the intensity level indicated on this chart. [Demon- 
stration] As you probably recognized, that was a phonograph 
record. The intensity to which it reached was about the intensity 
you would experience if you were near an aeroplane. 

We shall show you the possibilities of this system for utilizing the 
full intensity range. A good musical selection to show it is a march. 
The music will first be reproduced so softly that you will just be able 
to hear it near the threshold of hearing. Then the loudness will 
be increased through the entire intensity range of the system, and 
you will notice what range the extreme intensity level attains. The 
entire orchestra will be used. [Demonstration] 

In the next demonstration we shall use a steady source of sound, 
so that you may note the indicator reading of the intensity level and 
have time in which to compare it with the sound heard. The music, 
of course, varies in its intensity. The sound now will vary gradually, 
increasing in intensity to its maximum and then decreasing again. 
The sound is that of an electrical buzzer. When it is at its low in- 
tensity level you will recognize it. When it gets up to high intensity 
it will sound like something you have never heard before. I believe 
these few experiments will help you to realize what we mean by a 
wide intensity range. The march music probably illustrated it better 
than anything else. 

Now let us illustrate the fourth property that I mentioned: namely, 
that of controlling the sound while it is being produced as music. 
As most of you engineers know, at the present time the music that 
is sent out from broadcasting stations is controlled, in order to keep 
the intensity within the range that the system will transmit without 
distortion. The high-intensity sounds are lowered somewhat and 

328 HARVEY FLETCHER [j. s. M. p. E. 

the low sounds raised somewhat, in order to keep them within the 
working range. To show what we mean by such statements, we 
shall reproduce part of a musical selection under such control. The 
first time it will be reproduced just as the orchestra plays it; that is, 
with the same intensity. The second time we shall reproduce the 
same piece, but shall manipulate the controls so that the reproduction 
is kept within the very limited range of 30 decibels. The third 
time it will be played, the controls will be manipulated so that the 
contrasts will be very greatly emphasized, as you will notice on the 
chart. It will probably be magnified more than might be necessary 
for best musical taste. [Demonstration] 

Now, of course, it is difficult to judge by such a short musical 
selection, but I believe that it illustrates what we mean by electrical 
control of the music. In order to find out whether the enhanced 
volume range is worth anything musically, we should have to have 
a concert here and listen to some of the masterpieces played by a 
large symphony orchestra, and then have some musical critics listen 
to it while it is reproduced in these three different ways. I am sure 
there would be no uncertainty as to what the verdict would be, 
judging from the verdicts we have had in the past. 

In the next number we will use all the characteristics we have 
described, and all the artistry that we possess. Our soprano will 
sing again, with the orchestra. The orchestra will be reproduced 
through the two side channels, and the voice through the center 
channel. Also, it is arranged that the intensity of either center or 
sides can be placed at any desired level. [Demonstration] 

I see by the intensity level of the applause that you liked that about 
80 decibels worth. Now, of course, there may be musicians here 
who would say that no human voice can sing that loud and conse- 
quently that it is unnatural. That may be true, but the question 
is, did you like it? 

This really is the end of our program, but I am wondering if you 
wouldn't like to hear one more selection from the orchestra, using 
the three channels, that is, if you wish to stay. We shall now end 
the program with a rendition of the William Tell Overture, and this 
time again all three channels will be used. [Demonstration] 

I think that you will agree that we have demonstrated that the 
system is new in the following respects : It has auditory perspective, 
that is, it will reproduce the sound as though it were on the stage 
and coming from all positions on the stage; it reproduces a wider 


range of pitch than most other systems; it reproduces a wider in- 
tensity range than previous systems; and, of course, it has the novel 
feature, which may not be considered entirely new, of permitting 
the volume and the quality of the reproduced sound to be controlled. 
The loud speakers behind the screen reproduced all the sound that 
you heard. The lower part of each loud speaker reproduced the 
low frequencies, up to 300 cycles. The two loud speakers on the 
top of each unit reproduced the frequencies from 300 up to 15,000 
cycles. The low frequencies spread out all over the audience, as 
they have very little directional properties; but the high frequencies, 
as I have emphasized several times, are very directive. Therefore, 
they must be brought out of the loud speakers in such a way as to 
cover the entire hall; and that is why the horns have peculiar shapes. 
There is a section that points to each one of you from each of these 
units ; and they are so arranged that a spherical wave comes out of 
the front, and consequently there are no directional effects such as 
occur with most loud speakers. 


Filmcraft. A.Brunei. Geo. Newnes, Ltd., London, 1933; 238pp. 

Comparatively few books have been written on the technic of film production 
by experienced professional workers. This book should be welcomed, therefore, 
by the amateur cine enthusiast because it represents an outline of film produc- 
tion technic written by a well-known British director. Helpful information is 
included on all phases of the subject from the selection and preparation of the 
scenario to the final editing of the picture. The working staff is planned along 
lines similar to those of a professional studio. The text contains several ex- 
amples of actual scenarios and working scripts. An abbreviated glossary of 
technical terms used in film production is included. The hand-book closes with a 
number of short articles on various phases of film production written by experts 
at several of the British studios. Typical subjects treated are "Commercial 
Cutting," "Film Writing," "Lighting and Its Application," "Notes on Art Direc- 
tion," and "Notes on Direction." This little book should prove a useful addition 
to any library on film production, either amateur or professional. 


Film Technique. V. I. Pudovkin. Translated by I. Montagu. Geo. Newnes, 
Ltd., London, 2nd Edit., 1933; 204 pp. 

The author of this work on motion picture production technic is one of a school 
of Russian directors who have grown up within the past decade, and whose repu- 
tation has been based on their ability to make pictures that contain a flowing 
composition or rhythm. Editing or "montage" is considered the essence of film 
art, for by its subtle use scenes may be welded together smoothly and the tempo 
changed more or less at will. The first half of the book is divided into three essays, 
the first of which is a clear introduction to the other two. In the second part the 
principles of scenario construction are soundly treated in simple, understandable 
terms, with illustrative examples. The third part is a philosophical analysis of 
the process of motion picture production. The director must dominate the mak- 
ing of a picture and should follow through the cutting. Praise is repeatedly given 
the work of certain American directors, such as D. W. Griffith, but most American 
pictures are not considered representative of the author's ideas. 

Three new chapters have been added in the second edition, dealing respectively 
with "Close-ups in Time," "Asynchronism as a Principle of Sound Film," and 
"Rhythmic Problems in My First Sound Film." In the first of these the author 
describes his method of utilizing scenes or bits of scenes made with the ultra-rapid 
camera to incorporate "... various degrees of retarded speed of movement 
integrally in the construction of a given editing phase." To be most effective, 
sound must be edited into a film rather than recorded solely at the same time that 
the picture is taken. Several illustrations of this principle are given in connection 
with the Russian sound picture Deserter, directed by the author in 1933. A 
glossary of notes contains many useful comments. G. E. MATTHEWS 




At the regular meeting held at the Mid-West Film Company, at Chicago, 
April 12th, a symposium on sensitometry was held, as follows: 

"Sensitometry and the Laboratory Man," H. Anders, Jam Handy Picture 

Service, Inc., Chicago, 111. 
"Sensitometry and the Sound Man," J. Elliot Jenkins, Jenkins & Adair, 

Chicago, 111. 
"Sensitometry and the Cameraman," E. J. Cour, Jeencour Productions, 

Chicago, 111. 

A sensitometer built according to Eastman specifications was demonstrated. 
The meeting was well attended and great interest was shown in the proceedings, 
the subject of sensitometry having been chosen because of the many demands 
by members of the Section for a review of the practical applications of sensitome- 


At the meeting held at the Hotel Pennsylvania on April llth, members of the 
Amateur Cinema League and the Metropolitan Motion 'Picture Club were in- 
vited as guests of the section, to participate in a meeting that was devoted en- 
tirely to the interests of amateur cinematography. Despite the inclement 
weather, 110 persons attended the meeting, and evidenced their great interest in 
the proceedings by the enthusiasm shown toward the outstanding presentations 
by prominent amateur "filmers," as follows: 

Screening: Century of Progress, 1933, in Kodacolor, by H. H. Johnson, New 
York, N. Y. 

Slow Motion Diving Studies in Color, by E. Zacher, Hartford, Conn. 

A Christmas Story, in Kodacolor, by E. M. Barnard, Arkansas City, 

Chartres Cathedral and Venice, in Kodacolor, by J. V. Hansen, Washing- 
ton, D. C. 

Cinecoles Review, by R. Coles and C. Coles, Brooklyn, N. Y. 

Under the Maple Leaf, with special disk-sound accompaniment, by H. H. 
Jones, Buffalo, N. Y. 

Papers: "The Broader Aspects of Kodacolor," by F. A. Beach, Technical 
Consultant, Amateur Cinema League. 

"Development of 16-Mm. Projection in the Home," by R. C. Holslag, 
Movie Makers Magazine, New York, N. Y. 

Century of Progress (1933), Chartres Cathedral and Venice (1932), and Under 
the Maple Leaf (1932) were listed by Movie Makers magazine among the ten best 



amateur films of the years indicated. Slow Motion Diving in Color was given 
honorable mention in 1933. 

All those who participated in the program were members of the Amateur 
Cinema League. 


Two meetings were recently held by the Sub-Committee on Exchange Practice 
specifically for the purpose of studying the recently proposed 1700-foot reel length. 
Both meetings were held at the Great Northern Hotel, in New York, N. Y., the 
first on March 26th, and the second on April 2nd, and were well attended by not 
only the members of the Committee, but representatives of the various important 

As a result of the study made of the advantages and disadvantages of 
the 1700-foot length, the general consensus of opinion indicated disapproval of the 
1700-foot length, in view of its not being able to accomplish the objective desired 
in establishing that length ; namely, the elimination of doubling of reels by the 
projectionist, and the consequent waste of material and mutilation of the film. 

The Committee and exchange representatives were generally in favor of con- 
tinuing to use the 1000-foot reel length unless economic considerations should 
favor a 2000-foot length. In any event, lengths between 1000 and 2000 feet were 
objected to. 

A tentative report for presentation at the Spring Convention at Atlantic City 
was prepared, and was submitted on April 4th to the Projection Practice Com- 
mittee for its consideration. 


At a meeting held at the Paramount Building, New York, N. Y., on April 4th, 
the tentative report of the Sub-Committee on Exchange Practice dealing with 
the proposed 1700-foot reel length was carefully considered from the point of view 
of projection practice, with the result that the two Committees concurred in their 
general conclusions although they arrived at them from different considerations. 

The original proposal was to establish the length of 1700 feet as a maximum. 
As the film capacity of reels in common use in projection rooms is 3450 feet, 
the establishment of a 1700-foot maximum would not prevent doubling, particu- 
larly as it is common practice among projectionists to transfer the film to their 
own reels, which are generally in better condition than the reels supplied by the 

In order to prevent doubling, it would be necessary for the minimum length of 
film to be somewhat greater than half the capacity of the reel, which means that 
1750 feet should be the minimum, and about 2000 feet the average or nominal 
length. Such a length would make doubling impossible, and the objectives de- 
sired from making such a change would be accomplished. 

The Committee, although disfavoring the 1700-foot length, or any length 
between 1000 and 2000 feet for the reasons stated above, expressed its willingness 
to agree to whichever of the two latter lengths economic considerations might 
prove to favor. 

The report of the Committee was presented at the Spring Convention, at 
Atlantic City, and will be published shortly in the JOURNAL. 



At a meeting held at the General Office of the Society on April 6th, the revision 
of the Standards Booklet was carried to the point of preparing printer's proofs for 
the final inspection of all the members of the Committee before publication in the 
JOURNAL. Seventeen new charts have been added to the original fifteen, some 
of the latter being superseded by the new ones. Tolerances have been added 
throughout, both in the English and the metric systems, both composite and 
break-down drawings have been provided for film layouts, etc., and in many other 
respects the booklet has been made very complete and up-to-date. 

The final report of the Committee will be presented at the Spring Convention, 
Atlantic City, April 23-26, 1934. 


Results of the membership campaign thus far have been gratifying. Al- 
though the campaign did not begin until January 15th, at which time the reduc- 
tion of dues and other fees went into effect, 170 new members have been added 
to the roll since October 1st. The number of delinquent members this year is 
considerably smaller than the number last year, probably due also to the reduction 
of the fees, although many members have taken advantage of the reductions to 
apply for transfer to the higher grades. 

The special committee appointed by the Board of Governors to re-grade the 
membership will announce the results of its work at the Spring Convention. 
Members of the Society are urged to assist in continuing the campaign for new 
members as vigorously as it has started. 


The Society regrets to announce the death of Dr. Peter A. Snell, last year the 
holder of the S. M. P. E. Fellowship established through the generosity of the late 
George Eastman. Dr. Snell, 27, died March 14th, at Baltimore. He was a 
graduate of Hill Preparatory School and Princeton University, and received the 
degree of doctor of medicine from the University of Rochester, in 1933. While 
studying at Rochester he specialized in the physiology of vision, and pursued the 
investigations on visual fatigue that formed the subject of his S. M. P. E. Fellow- 
ship. His report was presented before the Society at the Fall, 1932, Meeting at 
New York, N. Y. 


ATLANTIC CITY, N. J., APRIL 23-26, 1934 

Approximately two hundred members and guests of the Society attended the 
various sessions of the Spring Convention at Atlantic City. The Convention 
opened on Monday morning with a general session, including reports of Com- 
mittees and a special meeting for Atlantic City projectionists, exhibitors, and 
managers. At the latter meeting short addresses were made by President Gold- 
smith, Mr. F. H. Richardson, and Mr. William Reed, who is perhaps the oldest 
projectionist in America, in point of length of service. 

At noon of the opening day, an informal luncheon was held for the members and 
guests. A short address of welcome was given by President Goldsmith, fol- 
lowed by addresses by Mr. Thomas Husselton, Secretary of the Atlantic City 
Chamber of Commerce, and Major William Casey, City Commissioner. 

The program of papers and presentations, as actually followed at the sessions, 
is presented herewith. At the semi-annual banquet, held on Wednesday evening, 
the members were addressed briefly by Mr. G. D. Lai, a member from India, and 
by Mr. Strickland Gilliland, humorist and author, of Washington, D. C. The 
principal address of the evening was presented by Mr. Sol A. Rosenblatt, Divi- 
sional Administrator, National Recovery Administration, who spoke on the 
various aspects of the motion picture code. Mr. Rosenblatt was appropriately 
introduced by President Goldsmith. 

Credit for the success of the Convention is largely due to the efforts of Mr. W. C. 
Kunzmann, Convention Vice-President, and Mr. J. O. Baker, Chairman of the 
Papers Committee. Others to whom credit is due were Mr. Harry Blumberg, 
Chairman of the Local Arrangements Committee; Mr. W. Whitmore, Chairman 
of the Publicity Committee; Mr. J. Frank, Jr., Chairman of the Apparatus Ex- 
hibit Committee; Mr. H. Griffin, in charge of projection; the officers and mem- 
bers of Atlantic City Local No. 310; Mrs. M. C. Batsel, Hostess; and Mr. J. 
Greenberg, Secretary of the Philadelphia Film Board of Trade. 

The sound and projection equipment used in the meetings and at the banquet 
was supplied and installed by the RCA Victor Company, the International Pro- 
jector Corporation, the Bausch & Lomb Optical Company, the National Carbon 
Company, and the National Theater Supply Company. 

Monday and Tuesday evenings were devoted to film programs as follows: A 
hand-synchronized 16-mm. travelogue, by Mr. Hamilton Jones, of Buffalo, N. Y.; 
Paramount News, Let's You and Him Fight, Paramount Pictures Corp.; Maid in 
Hollywood, Metro-Goldwyn-Mayer; Stand Up and Cheer, Fox Film Corp.; 
Twenty Million Sweethearts, Warner Bros. First National Pictures; Sisters under 
the Skin, Columbia Pictures Corp.; Beauty and the Beast, Vitagraph Corp.; The 
China Shop and Three Little Pigs, United Artists. 


The last two pictures were projected at the banquet; in addition, the banquet 
entertainment included a Hawaiian orchestra, through the courtesy of Mr. 
Richard Endicott, of the Steel Pier Theater, Atlantic City. The proceedings of 
the banquet, between 10:30 and 11:00 P.M., including the introductory remarks 
by President Goldsmith and the address by Mr. Sol A. Rosenblatt, were broadcast 
over the red network of the National Broadcasting Company. Passes to the 
various theaters of Atlantic City were kindly provided for the members by Mr. 
Coplan, Seashore Theaters, Inc.; Mr. Endicott, Steel Pier Theater; and Mr. H. 
Walters, Ventner Realty and Leasing Company. 


An interesting demonstration on Monday which aroused numerous comments 
was that given by C. E. Lane on the properties of wave filters. A series of 
pendulums attached to a beam were interconnected by metal spring bands. 
At certain frequencies the amplitude of the wave imparted mechanically to the 
first pendulum was gradually filtered out so that the last pendulum, was motion- 

The papers on 16-mm. equipment on Monday afternoon stimulated much 
discussion. This session proved to be one of the most interesting of the entire 
convention. Many favorable comments were heard relative to the improved 
sound quality on the 16-mm. films, particularly the quality of the reduction prints 
from 35-mm. feature pictures. The quality of the sound on the 16-mm. Koda- 
color was also of a high order. 

One of the most interesting demonstrations of amateur equipment was that 
given by H. Jones, who hand-synchronized over 50 different phonograph records 
with a travel picture on 16-mm. film. 

The combined reports of the Projection Practice and Exchange Practice Com- 
mittees, presented Tuesday afternoon, indicated that a decided stand would be 
taken opposing the adoption of a reel of 1700 feet length and favoring a 1000- or a 
2000-foot reel. Much evidence was advanced to show the fallacy of the inter- 
mediate size of reel. 

Dimmick and Belar's demonstration of the new twin triangular diaphragm 
sound track for noiseless recording represented, probably, the finest sound repro- 
duction of the Convention. The demonstration was given twice on Wednesday in 
order that those attending the Lighting Session could have the opportunity of 
listening to it. 

One of the liveliest discussions of the meeting occurred after B. Schlanger 
opened the forum Wednesday morning, on "What Is Wrong with the Shape of 
the Motion Picture?" The suggestions raised by Mr. Schlanger indicated much 
constructive thought on his part and his claim that the present picture includes 
only a small portion of the area covered by natural vision elicited many comments. 
It was evident from the resulting discussion that the wide-film pictures shown in 
1930-31 approached more nearly Mr. Schlanger's ideal shape and were more 
realistic than present-day films. 

The use of the piezoelectric properties of quartz, tourmaline, and other crys- 
talline substances to construct extremely accurate "time clocks" for use hi radio 
broadcasting stations and sound studios was explained in a paper by F. R. Lack, of 

336 SPRING CONVENTION [j. s. M. P. E. 

the Bell Telephone Laboratories, on Thursday morning. A practical application 
of the use of piezoelectric crystals was given during the apparatus symposium 
by A. L. Williams who demonstrated a microphone which utilized a crystal of 
Rochelle salt. 

Originals and duplicates made by the British Dufaycolor process on 35-mm. 
and 16-mm. films were shown by W. H. Carson. This is a three-color additive 
process using a line screen which is coated on the film. Generally favorable 
comment was heard on the beauty of the colors, which were of pleasing pastel 



Morning: General Session 

"Technical ^Committees Their Organization and Policies," L. A. Jones, 

Engineering Vice-President. 

Report of the Progress Committee, J. G. Frayne, Chairman. 
Report of the Committee on Standards and Nomenclature, M. C. Batsel, Chairman. 
"Some Early Experiments in Photographic and Motion Picture Work," F. E. 

Ives, Philadelphia, Pa. 

"Oscilloscope," H. F. Mallina, Bell Telephone Laboratories, New York, N. Y. 
"History of Sound Pictures," W. E. Theisen, Honorary Curator, Los Angeles 

Museum, Motion Picture Division, Los Angeles, Calif. 

Afternoon: Amateur and 16-Mm. Session 

"A Demonstration of the Properties of Wave Filters," C. E. Lane, Bell Telephone 
Laboratories, New York, N. Y. 

"A Sixteen-Millimeter Bound Camera," G. L. Dimmick, C. N. Batsel, and L. T. 
Sachtleben, RCA Victor Company, Camden, N. J. 

"Sixteen-Millimeter Sound Motion Pictures in Color," C. N. Batsel and L. T. 
Sachtleben, RCA Victor Company, Camden, N. J. 

"Recent Examples of 16-Mm. Sound Pictures on Double Sprocket Hole Film," 
A. W. Carpenter, H. J. Hasbrouck, J. F. Nielsen, and E. R. Ross, United Re- 
search Corporation, Long Island City, N. Y. 

Report of the Committee on Non-Theatrical Equipment, R. F. Mitchell, Chair- 

"Problems of the Amateur Motion Picture Maker," R. C. Holslag, Amateur 
Cinema League, New York, N. Y. 


Morning: Projection Session 

"Factors Covering the Design of Projection Lamps, and Their Application to 

Equipments," F. E. Carlson, General Electric Company, Cleveland, Ohio. 
"Operating Characteristics of the High-Intensity A-C Arc for Motion Picture 


Projection," D. B. Joy and E. R. Geib, National Carbon Company, Cleve- 
land, Ohio. 

"The Relationship of the High-Intensity A-C Arc to the Light on the Projection 
Screen," D. B. Joy and E. R. Geib, National Carbon Company, Cleveland, 

"A-C Adapters for Low-Intensity Reflecting Arc Lamps," R. Miehling, New 
York, N. Y. 

"Effect of Aperture Lenses on the Illumination of Motion Picture Screens," 
W. B. Ray ton, Bausch & Lomb Optical Company, Rochester, N. Y. 

Afternoon: Exchange and Theater Session 

"Simple Theory of Three-Element Vacuum Tubes," H. A. Pidgeon, Bell Tele- 
phone Laboratories, New York, N. Y. 

Report of the Sub-Committee on Exchange Practice, T. Faulkner, Chairman. 

Report of the Projection Practice Committee, H. Rubin, Chairman. 

"Reel Problems in Exchange Practice," T. Faulkner, S. M. Chemical Company, 
New York, N. Y. 

"Technical Aspects of Theater Operation," R. M. Wilcox and L. W. Conrow, 
Electrical Research Products, Inc., New York, N. Y. 

"Cheapness Does Not Always Pay," F. H. Richardson, New York, N. Y. 

"The Motion Picture Theater Auditorium," B. Schlanger, New York, N. Y. 


Morning: Sound Session 

"Some Recent Improvements in Equipment and Technic in the Production of 
Motion Pictures," E. A. Wolcott, RKO Studios, Hollywood, Calif. 

"Some Considerations in the Design of Sound Reproducing Equipment," F. 
Marion and G. Friedl, Electrical Research Products, Inc., New York, N. Y. 

"On the Realistic Reproduction of Sound, with Particular Reference to Sound 
Motion Pictures," H. F. Olson and F. Massa, RCA Victor Company, Cam- 
den, N. J. 

Report of the Sound Committee, L. W. Davee, Chairman. 

"An Improved Sound System for Noiseless Recording," G. L. Dimmick and H. 
Belar, RCA Victor Company, Camden, N. J. 

"Recent Optical Improvements in Western Electric Sound Film Recording 
Equipment," W. Herriott and L. B. Foster, Bell Telephone Laboratories, 
New York, N. Y. 

"Care and Operation of Theater Sound Systems," J. S. Ward and P. T. Sheridan, 
Electrical Products, Inc., New York, N. Y. 

Morning: Lighting Session 

"Studio Lighting," S. W. Woodside, Westinghouse Lamp Company, Bloomfield, 

N. J. 
"The Application of the Bi-Plane Filament Light Source to Spotlighting Service," 

G. T. Mili, Westinghouse Lamp Company, Bloomfield, N. J. 
Open Forum: "What Is Wrong with the Shape of the Motion Picture?" "How 

Can the S. M. P. E. Be of Better Service to the Industry?" 


"Developments in Spotlighting," H. Kliegl, Kliegl Bros. Stagelighting Company, 
New York, N. Y. 

"Thyratrons and Their Application," E. H. Alexander, General Electric Com- 
pany, Cleveland, Ohio. 

"Stroboscopic Light High-Speed Photography," H. E. Edgerton and H. Germes- 
hausen, Massachusetts Institute of Technology, Cambridge, Mass. 


Morning: Laboratory Session 

"Continuous Optical Reduction Printing," A. F. Victor, New York, N. Y. 

"A Non-Slip Sound Printer," C. N. Batsel, RCA Victor Company, Camden, N. J. 

"An Optical Reduction Sound Printer," C. N. Batsel and L. T. Sachtleben, RCA 
Victor Company, Camden, N. J. 

"Properties of Piezoelectric Crystals," F. R. Lack, Bell Telephone Laboratories, 
New York, N. Y. 

"The English Dufaycolor Film Process," W. H. Carson, New York, N. Y. 

Open Forum: "Suggestions for Improvements in Motion Picture Laboratory 
Practice," "Possible Motion Picture Applications of the Principle of Audi- 
tory Perspective." 

"A Small Developing Machine," H. R. Kossman, Andre Debrie, Inc., New York, 
N. Y. 

Afternoon: Photographic Session 

"Piezoelectric Microphones," A. L. Williams, Brush Development Company, 

New York, N. Y. 
"Equipment for the Hard-of -Hearing," D. D. Halpin, Sonotone Corp., New York, 

N. Y. 

"Some Properties of New Agfa 35-Mm. Film," P. Arnold, Agfa Ansco Corpora- 
tion, Binghamton, N. Y. 
"The Failure of the Reciprocity Law in Photographic Exposure," L. A. Jones 

and J. H. Webb, Eastman Kodak Company, Rochester, N. Y. 
"A Year's Practical Experience with a High-Speed Timing Camera," H. T. Day, 

Electrical Research Products, Inc., New York, N. Y., and F. Tuttle, Eastman 

Kodak Company, Rochester, N. Y. 
"The Microdensitometer as a Laboratory Measuring Tool," W. R. Goehner, Bell 

Telephone Laboratories, New York, N. Y. 
"A Sweep Oscillator Method of Securing Wide Band Frequency Response Spectra 

on Short Lengths of Motion Picture Film," J. Crabtree, Bell Telephone 

Laboratories, New York, N. Y. 




Volume XXII JUNE, 1934 Number 6 



Progress in the Motion Picture Industry: Report of the 
Progress Committee 341 

Report of the Projection Practice Committee 379 

Report of the Sub-Committee on Exchange Practice 386 

Society Announcements 391 

Author Index, Volume XXII (January to June, 1934) 393 

Classified Index, Volume XXII (January to June, 1934) 395 





Board of Editors 
J. I. CRABTREE, Chairman 



Subscription to non-members, $8.00 per annum; to members, $5.00 per annum, 
included in their annual membership dues; single copies, $1.00. A discount 
on subscriptions or single copies of 15 per cent is allowed to accredited agencies. 
Order from the Society of Motion Picture Engineers, Inc., 20th and Northampton 
Sts., Easton, Pa., or Hotel Pennsylvania, New York, N. Y. 

Published monthly at Easton, Pa., by the Society of Motion Picture Engineers. 

Publication Office, 20th & Northampton Sts., Easton, Pa. 
General and Editorial Office, Hotel Pennsylvania, New York, N. Y. 
Entered as second class matter January 15, 1930, at the Post Office at Easton, 
Pa., under the Act of March 3, 1879. Copyrighted, 1934, by the Society of 
Motion Picture Engineers, Inc. 

Papers appearing in this Journal may be reprinted, abstracted, or abridged 
provided credit is given to the Journal of the Society of Motion Picture Engineers 
and to the author, or authors, of the papers in question. Exact reference as to 
the volume, number, and page of the Journal must be given. The Society is 
not responsible for statements made by authors. 

Officers of the Society 

President: ALFRED N. GOLDSMITH, 444 Madison Ave., New York, N. Y. 
Executive Vice-P resident: HAROLD C. SILENT, 7046 Hollywood Blvd., Los 

Angeles, Calif. 

Engineering Vice-President: LOYD A. JONES, Kodak Park, Rochester, N. Y. 
Editorial Vice-President: JOHN I. CRABTREE, Kodak Park, Rochester, N. Y. 
Financial Vice-President: OMER M. GLUNT, 463 West St., New York, N. Y. 
Convention Vice-President: WILLIAM C. KUNZMANN, Box 400, Cleveland, Ohio. 
Secretary: JOHN H. KURLANDER, 2 Clearfield Ave., Bloomfield, N. J. 
Treasurer: TIMOTHY E. SHEA, 463 West St., New York, N. Y. 


EUGENE COUR, 1029 S. Wabash Ave., Chicago, 111. 
HERFORD T. COWLING, 7510 N. Ashland Ave., Chicago, 111. 
ARTHUR S. DICKINSON, 28 W. 44th St., New York, N. Y. 
RALPH E. FARNHAM, Nela Park, Cleveland, Ohio. 
HERBERT GRIFFIN, 90 Gold St., New York, N. Y. 
EMERY HUSE, 6706 Santa Monica Blvd., Hollywood, Calif. 
WILBUR B. RAYTON, 635 St. Paul St.. Rochester, N. Y. 
HOMER G. TASKBR, 41-39 38th St., Long Island City, N. Y. 


Summary. This report of the Progress Committee covers the period June, 1933, 
to April, 1934. The advances in the cinematographic art are classified as follows: 
(/) Cinematography, (II) Sound Recording, (III} Sound and Picture Repro- 
duction, (IV) Film Laboratory Practice, ( V) Applications of Motion Pictures, 
( VI) Publications and New Books, ( VII) Appendix. 

In preparing the report on progress in the motion picture industry 
for the year 1933, the Committee appeared about to undertake what 
might be termed a very disheartening task. The members of the 
Committee who have been in close touch with the producing phase of 
the industry took the rather pessimistic point of view that little, if 
any, progress could be reported for the past year. The economic 
difficulties in which most of the major producers found themselves 
during the past year, coupled with the bank holiday and an acute 
labor crisis during the summer, prevented the studios from giving as 
much attention to technical improvements and advances as they 
might otherwise have done. 

In view of this situation it was a matter of much gratification to 
the Committee to find considerable evidence of progress in new 
equipment in the various reports submitted for consideration. It is 
true that there is little to report in the way of new films and emul- 
sions either for photographic or sound recording purposes a situa- 
tion that can undoubtedly be attributed to the fact that during the 
preceding years great advances had been made in that branch of the 

There is little to report in the way of new professional cameras and 
accessories during the year, but there are many items of interest 
connected with improvements in studio illumination. Of particular 
interest is the new type of mirror coating employing aluminum, 
which has received considerable publicity during the past year. 

In the field of color there has been little to report, except that 
the public appears to be evincing more interest in colored cartoons 

* Presented at the Spring, 1934, Meeting at Atlantic City, N. J. 


342 PROGRESS COMMITTEE f j. s. M. P. E. 

and the stage appears to be set for an impressive revival of color in 
important feature pictures. 

There are many more new developments in amateur cinematog- 
raphy than in the professional field. Many new cameras have been 
presented to the public by manufacturers in this country as well as in 
Germany and Great Britain. 

In the field of sound recording there are many evidences of progress. 
The wide-range and high-fidelity systems are gradually replacing the 
older recording systems in studios. One of the major electrical cor- 
porations, during the past year, has announced a device for making 
the recording valve used by it more nearly perfect in its operation. 
Announcements have also been made by the electrical corporations of 
extremely portable light-weight, single-film recording systems ; and 
while they are primarily intended for use in newsreel work, there is 
some possibility of introducing the corresponding recording technic 
into the studios in place of the standard dual systems now commonly 

A number of new sound recording accessories are reported, among 
which are wave analyzers for facilitating measurements of harmonic 
distortion in audio-frequency circuits, as well as noise meters, which 
have a very timely application in sound recording and in measuring 
camera noise on sets. The report this year mentions a new item not 
heretofore reported: namely, a complete equipment for recording 
sound on 16-mm. film, using an amateur camera. It will be interest- 
ing to follow the reception that will be accorded by the public to this 

In the field of sound and picture production is an announcement 
from Germany of several new types of projectors having many novel 
features. In the United States there have not been many items 
offered in this field, but it has been reported to the Committee that a 
considerable number of theaters in the country are bringing their 
sound reproducing equipments up to date. 

There is little to report on film laboratory practice, but there are 
evidences here and there of general improvement in the film develop- 
ing situation. The quality of release prints still remains question- 
able, and the Committee regrets that it has nothing definite to 
report in the matter of improvement in that vital field. 

As to the applications of motion pictures, considerable progress 
appears to have been made in the practical application of a small 
light-weight, high-speed camera capable of taking 2000 frames per 


second on the standard 16-mm. film. This camera is associated with 
a timing device, which should make it an important tool in all kinds 
of accurate time analyses. 

In the Appendix of the report is listed an account of progress in 
motion pictures in Great Britain and Japan. It is of interest to note 
that the motion picture industry in Great Britain has made con- 
siderable progress during the past year, as evidenced by the ex- 
cellent reception given in this country to sound pictures made there. 

The Committee wishes to thank the following firms for supplying 
photographs for use in this report: Bell & .Howell Company; 
General Radio Company; Eastman Kodak Company; Electrical 
Research Products, Inc. ; Paramount Productions, Inc. ; RCA Victor 
Company; Victor Animatograph Co.; and Western Electric Com- 
pany, Ltd. 

J. G. FRAYNE, Chairman 






A . Professional 

1. Films and Emulsions 

2. Cameras and Accessories 

3. Studio Illumination 

4. Color 

B. Amateur Cinematography 

1. General 

2. Cameras 

3. Projectors 

4. Color 


A . Professional 

1. General 

2. Recording Equipment (Dual System) 

3. Recording Equipment (Single System) 

4. Accessories 

B. Amateur 



1. Sound Equipment 

2. Projectors and Accessories 

3. 16-Mm. Sound-on-Film 


1. Film Development 

2. Laboratory Equipment 


1 . Education 

2. Timing Devices 

3. Miscellaneous 



A . General Field of Progress of the Motion Picture Industry in Great 


B. Progress in the Motion Picture Industry in Japan 


A . Professional 

1. Films and Emulsions. A review of the literature for the past 
year reveals very little new information concerning motion picture 
film emulsions. Extensive use has been made, however, of the fast 
panchromatic materials since their introduction three years ago. 
Several additions were made to the list of film emulsions available to 
the still cameraman. Panchromatic roll films of high speed and 
color-sensitiveness combined with fine grain were introduced. One 
of these films was somewhat similar in speed, color-sensitiveness, and 
grain characteristics to the motion picture background negative film 
introduced in 1932. l 

In the sound recording emulsion field the Agfa Ansco Corporation 
has announced the TF III 35-mm. motion picture film for sound 
recording with the variable density method. Its speed, about five 
times greater than that of positive, the long straight-line density 
curve, the low gamma infinity, and its great resolving power are 
stated to qualify this film as a high-class dependable recording 

This corporation announces also the TF IV 35-mm. motion picture 
film for the variable width recording method. It is claimed to be 
superior in contrast and resolving power to positive film, and about 
2 to 3 times the speed of the latter. 

In the realm of theory, a number of interesting papers have ap- 


peared that deal with emulsion research. Methods of testing gela- 
tins for photographic use were described by Fuchs 2 who stated that 
the emulsion maker's experience and intuition still are important 
factors in governing the selection of gelatins. Heyne 3 reviewed the 
history of gelatin and pointed out that we still have not attained the 
ideal position of being able to produce a particular emulsion from an 
absolutely inactive gelatin by predetermined additions of one or more 
sensitizers. Some factors in the preparation of photographic emul- 
sions were discussed by Charriou, 4 such as the concentration of gelatin 
during precipitation, the rate of precipitation, the effect of tempera- 
ture and concentration of silver nitrate during preparation. 

The interesting research on photographic emulsions by Carroll 
and his co-workers at the U. S. Bureau of Standards was extended. 
In conjunction with Hubbard, 5 data were published dealing with 
the mechanism of hypersensitization and with sensitization by sodium 

Russian chemists published a group of papers dealing with several 
aspects of photographic emulsion manufacture. 6 The emulsion ripen- 
ing process and also the chemical sensitizing of the finished emul- 
sion showed a displacement of spectral sensitivity toward the longer 
wavelengths in confirmation of Sheppard's theory of sensitizing 
nuclei. It was observed that the crystals of Russian emulsions were 
more cubical in form than crystals of emulsions made elsewhere. 

Patent protection was granted several applicants for methods of 
halation prevention. 7 Capstaff 8 would obtain a tinted film with a 
clear sound track area by tinting the entire support surface and then 
removing the layer containing the dyes along the sound track. 
Crouch 9 described a method of preventing buckling of the film along 
the sound track by treating it with water and a solvent above normal 

The usual large number of patents were issued to various persons 
on the subject of improvements in cellulose compositions, which in- 
dicated that research has been going forward in that important field. 
The reclamation of nitrate and acetate supports is a subject of 
fundamental interest, and two methods for such recovery were dis- 
closed. 10 

Two unusual patents proposed the incorporation of magnetic 
particles in an adhesive material coated on a film strip, the purpose 
being to permit the recording of sound on the magnetic particles. 11 
Several methods were disclosed for producing non-inflammable 

346 PROGRESS COMMITTEE [j. s. M. p. E. 

motion picture film; one type comprising a layer of insoluble gelatin 
coated on each side with successive layers of rubber and a pro- 
tective varnish. 12 

2. Cameras and Accessories. The year 1933 has offered little in 
new emulsions and cameras in the professional field. Announce- 
ments of more compact and silent cameras in the report of last year 
have not been followed by commercial production or by the intro- 
duction of this type of camera into general studio use. In the 
United States, at least, the standard 35-mm. camera in a blimp hous- 
ing continues to be used in the great majority of cases. Improved 
silent cameras are undoubtedly on the way, but are not yet a reality so 
far as adoption by the industry is concerned. 

In the line of accessories, the Bell & Howell Company has an- 
nounced a sunshade of conventional design, but adapted for lenses of 
very short focal length (241). Provision is made for using filters, 
diffusing disks, etc. lz 

An account is given in the American Cinematographer of a camera 
carriage with a crane arm adjustable for any camera height from 26 
inches to 6 x /2 feet, equipped with conventional panoram and tilt 
head and carried on four rubber- tired wheels. 14 

An interesting hydraulic camera dolly has three wheels, electric 
motor drive, and automatic cable reels. 15 The hydraulic elevating 
column has an elevation range from 18 to 66 inches. 

A number of new ultra-rapid lenses have been announced during 
the past year. Zeiss has produced an //0.85 objective especially de- 
signed for x-ray cinematography. 16 The Astro-Gesellschaft offer 
//0.95 lenses in focal lengths of 52 mm. and 75 mm. for standard film, 
and a 35-mm. outfit for amateur use. 17 The Pantar objective, with 
an aperture ratio of // 1.0, has also been described, 18 and there have 
been an unusual number of patents relating to the details of objective 
construction. 19 

The German Askania Works have completed during the year the 
development of a telephoto lens that is new in its application to cine- 
matography. This objective is described by H. Acht and F. Beck 20 
in an article containing excellent illustrations of the construction of 
the lens and the results that can be obtained with it. It employs 
reflectors instead of refracting elements, like an astronomical tele- 
scope, and this type of construction appears to have many advantages 
when extremely long focal lengths are required. 

H. Naumann has published a paper on the history and characteris- 

June, 1934] 



tics of variable focus lenses now used on motion picture cameras. 21 
There is a growing interest in the type of photography that employs 
light at the infrared end of the spectrum, as shown by the number of 
published articles on the subject. 22 A motion picture was made by 
infra-red radiation of subjects in complete darkness on Oct. 9, 1933, 
at the Gaumont-British Theater, London, England 220 (Fig. 1). 

A series of articles has just been published by W. Taylor 23 describ- 
ing the methods and machinery used in the optical shop of an English 
manufacturer; the author seems to consider his methods as the 
most modern ones. There have appeared a few articles on the depth- 
of-field of camera objectives, one by A. A. DeBois 24 and one by J. F. 

FIG. 1 . Motion picture made with infra-red radiation. Radiation supplied 
by two 2000-watt spotlights covered with infra-red niters. Vinten camera 
with //1. 9 lens used. Pictures exposed 16 per second. Hypersensitized 
infra-red film. (Illustrated London News} 

Westerberg, which was published in the JOURNAL. 25 Westerberg has 
also published in the International Photographer 26 a series of tables 
that seem to cover all the optical and photographic data that could 
ever be required by a cameraman. 

3. Studio Illumination. About the time of the 1933 Spring Meet- 
ing the two major lamp companies introduced a complete line of 
projection lamps, including lamps suitable for the recently intro- 
duced 8-mm. film projectors as well as the more familiar 16- and 35- 
mm. portable equipments. An interesting feature of the newer 
lamps is the adoption of a 25-hour life in order to gain higher screen 
brilliance; and the increase in wattage without proportionate in- 
crease of bulb volume, thus necessitating a high degree of forced 


ventilation with equipments employing them. Brief mention is 
made of these lamps in the report of the Non-Theatrical Equipment 
Committee presented at the April, 1933, Meeting. 

At this same time one of the well-known equipment manufacturers 
made available a new type of cored carbon especially applicable to 
a-c. power supply. This carbon operates at the unusually low volt- 
age of from 27 to 30 volts, and produces a bluish white light 
characteristic of the high-intensity flame carbons. This carbon 
should find its greatest application in the smaller theaters now using 
the low-power reflector arcs. 

The past year has witnessed the quite general adoption of what 
is now known as the No. 20 Photoflash lamp by many professional 
and amateur photographers. This lamp is of interest to the motion 
picture industry because of its general use by the still photographers 
employed by the producers, and its frequent use in movies bringing 
in the newspaper theme. In addition, two new lamps have been 
introduced; one known as the No. 10, of one-half the light output of 
the No. 20. This lamp is intended for amateur use and sells for a 
lower price. The second new lamp, known as the No. 75, has more 
than three times the light output of the No. 20 and is intended for 
newspaper photography covering large areas as well as for color 

Of special interest to motion picture cameramen has been the new 
Movieflood lamp, designed especially for motion picture photography 
in color. This lamp is rated at 2000 watts, and is photographically 
equivalent to about four and one-half 1000- watt general service 
lamps. It operates at a very high efficiency, which makes its light 
much richer in blue and violet radiation than the more familiar 
lamps. It is supplied in the PS-52 bulb, and has a life of 15 hours. 
A more detailed account of it and its uses was presented before the 
Pacific Coast Section of the S. M. P. E. and published in the July, 
1933, issue of the JOURNAL. 

On March 1, 1934, there was introduced a new larger sized Photo- 
flood lamp having four times the light output of the original 
Photoflood lamp. This new lamp is known as Photoflood No. 4, and 
the smaller original one as Photoflood No. 1. It has a consumption 
of about 1000 watts, and provides approximately 2 l / 2 times as much 
light as any other lamp of equal wattage. It is of particular interest 
to motion picture cameramen for close-up work as well as to the 
commercial photographer. 


Of unusual interest to the manufacturers of motion picture light- 
ing equipment has been the announcement by one of the large alumi- 
num producers of a method of treating aluminum to obtain very 
high reflection efficiency. Ordinary aluminum surfaces, matte or 
polished, have a reflection efficiency of 65 to 70 per cent, whereas the 
new surface ranges from 85 to 90 per cent. Aluminum surfaces rang- 
ing from highly specular to totally diffuse are obtainable by this 
treatment, and are especially resistant to atmospheric corrosion. 

An account of the use of aluminum in reflectors was presented 
before the Cleveland Section of the Illumination Engineering Society 
in January, 1934. It was explained by R. B. Mason, of the Alumi- 
num Research Laboratories, that a new type of anodic treatment 
known as "electrolytic brightening" gives the aluminum its 
brightened surface. By this process, it is claimed, an aluminum 
polished surface having a total reflectivity of 74 per cent had a 
reflectivity of 87 per cent after being subjected to the anodic treat- 
ment. It was further pointed out that after the electrolytic 
brightening operation a substantial oxide coating could be produced 
by anodic processes without any substantial loss of reflectivity. As a 
final step in producing the finished reflector, known as the Alray, the 
oxide coated surface is sealed to make it impervious to corrosive in- 

Another type of reflecting surface involving aluminum received 
considerable publicity in the scientific press last year. Aluminum 
metal is evaporated in vacuo on to the metal or glass mirror, producing 
thereon a reflecting surface of high efficiency, which is claimed to be 
free from any staining due to exposure or handling. This method 
has been described by Strong 27 and Edwards, 28 and is now being used 
for coating astronomical mirrors. It has not yet made its appear- 
ance in commercial studio equipment although several studios are 
experimenting with it. 

Kliegl Bros., of New York, have recently introduced a new type of 
incandescent lamp spotlight which employs a rhodium plated ellipti- 
cal reflector. This reflector, in conjunction with a suitable project- 
ing lens, allows an accurate control of the light over a wide angle. 
A spotlight involving this principle has produced more than three 
times the amount of illumination as compared with the more usual 
lens spots. In addition, it provides a very uniform, sharply defined 

The General Electric Vapor Lamp Company reports that a new 

350 PROGRESS COMMITTEE [j. s. M. p. E. 

type of high-efficiency, limited-pressure mercury arc is well along in 
development. This arc has an efficiency of about 40 lumens per watt, 
and is quite compact, a 425-watt unit being contained in a tubular 
lamp bulb 2 inches in diameter and 13 inches in over-all length. The 
light source itself is 6 inches long and about J /2 inch in diameter. 
The arc can be used for any of the studio applications for which the 
older type of low-pressure mercury arc has been used, and has the 
advantage of possible use with reflectors in broad-beam general light- 
ing units. The color quality of the light is considerably whiter than 
that of the older low-pressure form of the mercury arc, there being 
additional red and green light of value in panchromatic work. 

4. Color Cinematography. The interest in the use of color in 
professional cinematography noted in the previous progress report has 
been carried forward into the current year. The strengthened com- 
petition, with which the motion picture industry finds itself faced, 
as a consequence of advanced technic in radio presentation, the 
rehabilitation of the dramatic stage, and various other amusement 
facilities, has made it necessary for the industry to seek additional 
appeal for its productions. That one of those appeals should be 
color is a natural conclusion; that advances in the development 
of a satisfactory three-color process were available to supply that ap- 
peal was not coincidence, but the result of far-sighted planning by 
organizations such as Technicolor. 

The outstanding development during the current year has been 
the veritible craze for color cartoons, the outstanding example of 
which was Disney's Three Little Pigs, photographed in Technicolor's 
three-color process, establishing a new technic, and breaking records 
on every front. This successful excursion was followed by notable 
dramatic sequences M-G-M's The Cat and the Fiddle and 20th Cen- 
tury's The House of Rothschild. Cinecolor has also contributed to the 
cartoon field with several successful numbers. A bi-pack process 
was introduced by a British firm which utilized a double-coated film 
for printing. 28 " 1 

A large number of patents were issued covering improvements in 
35-mm. lenticulated films, particularly methods of printing such 
color records, but no extensive commercial use was known to have 
been made of these films except their application to small still cameras 
such as the Leica and Contax. Records made with these cameras 
would presumably be projected similarly to slide films. 286 

A German patent has been issued describing the making of a line- 


screen for color photography by using several spinneret tubes to form 
the lines in uniform succession on the carrier base. 29 A method of 
producing multicolored pictures was patented, consisting in the 
use of an impregnated support composed of esters of leuco com- 
pounds of a vat dye. This support was recommended to be exposed 
behind a color-separation negative to yield the positive, from which 
the unchanged leuco ester was to be washed out. 30 Several addi- 
tional patents were noted which dealt with various improvements 
in the additive color process using lenticulated film. 31 

B. Amateur Cinematography 

1. General. The past year has been marked by improvements of 
existing apparatus, materials, and processes, rather than by develop- 
ment of any new principles. Many reversible emulsions have been 
introduced in both the United States and Europe. These aim mostly 
at progress in speed, freedom from halation, good color-sensitivity, 
and fine grain. 32 - 33 

2. Cameras. Several new amateur cameras have been either 
introduced or improved during the past year. Features of the 
Cine-Kodak Special include a dissolve shutter, a turret carrying any 
two of a series of interchangeable lenses, a reflex focusing finder, and 
a spring motor with long uniform run characteristics. The speed is 
controlled from 8 to 64 frames per second, and interchangeable 
film chambers are provided having capacities of 100 and 200 feet. 
Other features included are a single-exposure trip, single-frame per 
turn and eight-frame per turn hand-crank or motor-drive shafts, 
and masks for double exposure and effect photography. 

The Bell & Howell Company has announced a model 70 E Filmo 
16-mm. camera. It is similar to the Filmo 70 D model with the ex- 
ception that the three-lens turrent has been eliminated. However, 
interchangeability with all types of lenses mounted for the Filmo 70 D 
is possible. The 70 E model can be operated at speeds from 8 to 64 
picture frames per second, and can be used for black-and-white or 
Kodacolor pictures (Fig. 2). 

The Bell & Howell Company also has announced a new model 
16-mm. camera of extreme compactness. There are two finders, one a 
direct sight and the other a waist level type. It can be operated at a 
speed of either 16 or 24 picture frames per second. The shutter is of a 
unique type with oscillating body action akin to that of a focal plane 
shutter. Its film capacity is 50 feet to the cartridge magazine 



[J. S. M. P. E. 

type, greatly simplifying the loading. A variety of lenses of various 
focal lengths and apertures is available, and the camera can be used 
for black-and-white as well as for Kodacolor pictures (Fig. 2). 
Of interest to amateurs is the 1934 model 3 Victor camera which has 
five operating speeds. Other features include duplex twin mounted 
spring motor, attached winding crank, built-in exposure guide, 
multiple view finder, etc. (Fig. 2). 

In Germany the Siemens A. G. have completed their entire 16-mm. 
program by producing the Siemens Rino-Kameras C and D. In 

FIG. 2. (^4) 16-Mm. camera (model 3, Victor Animatograph Corp.}; (B) 
16-mm. camera (model 121, Bell & How ell Co.); (C) 16-mm. camera (model 
70-E, Bell & Howell Co.). 

general the type D Siemens corresponds with the Siemens Kino- 
Kamera C, 34 with the exception that the latter has only one //1. 5 lens; 
whereas the model D is fitted with three lenses on a vertical slide, 
with apertures ranging from //1. 5 to//4.5 in different focal lengths. 
In the course of the past year, the firm of E. Leitz, at Wetzlar, 
Germany, produced several new lenses especially suitable for sub- 
standard film. In view of the improvements that have been made in 
the fine-grain emulsions, all their lenses were newly calculated. This 
is particularly true with two Leitz lenses, the Dygonf/2.8 and//3.5, 


both of which have a focal length of 20 mm., and even more so with 
the four Hektor lenses //I. 4/25-mm.,//2.5/35-mm.,//1.9/50-mm., and 
//2.5/50-mm., as well as with the Tele-Lens Telyt, which has a focal 
length of 75 mm. and a relative aperture of //4. All these lenses are 
obtainable in focusing mounting, and the Dygon //3.5/20-mm. can 
also be had in Fixfocus mounting. 

The firm of Jos. Schneider & Co., Kreuznach, has also been very 
active. Their Xenonf/1.3, 1.5 lens can be used in conjunction with a 
special lens, and is also particularly suitable for lenticulated film, such 
as Kodacolor and Agfacolor. With this lens a careful correction of 
the coma is effected. Schneider states that the Gauss type of lens is 
particularly suitable for such correction, as it makes this possible with- 
out affecting the elimination of other lens defects. 

3. Projectors. Continued improvement in 16-mm. projectors is 
evidenced by the announcements of the manufacturers. From Ger- 
many comes the announcement that the Grossraum projector has an 
unusual brilliancy, as it is equipped with a 75-volt, 375-watt lamp, 
and is therefore suitable for use in large auditoriums. It is provided 
with a new type beater movement. The advertising projector is 
fitted with a relay which effects the rewinding of the film at regular 
intervals. Consequently, the projection is not continuous, but re- 

During 1933 the firm of Zeiss Ikon produced a portable 16-mm. 
projector of great efficiency, the so-called Schmalfilm-Kinox. This 
apparatus is suitable for large projection and is built in the form of a 
suitcase. It is also said to be adapted to sound film projection. 

The firm of Lytax in Germany has produced a new projector called 
the Piccolo, with which either a 33-volt, 100-watt lamp, or a 100-volt, 
400-watt biplane lamp can be used, and thus very brilliant pictures 
can be projected. This apparatus incorporates a new movement 
constructed on the same lines as the Geneva cross. With this it has 
been possible to achieve a correct movement of the film within 60 
degrees of the rotary shutter. It is claimed that, due to the fact that 
the sprockets of the intermittent wheel are always in contact with 
several perforation holes, the durability and steadiness of the pro- 
jected pictures are both improved, and that by this method the 
apparatus is rendered suitable for Ozaphan film as well as for the usual 
safety film. 

4. Color. A new type of prism for exposing two images on a 
single frame of 16-mm. film was demonstrated at a meeting of the 


British Physical Society in February. The prism is said to split 
the beam from a single lens without loss of light or definition. The 
prism holder and the twin-lens mount can be interchanged with the 
ordinary lens in a few minutes. A projector fitted with a double- 
lens system is used to show the prints. 350 


A . Professional 

1. General. There has been no outstanding alteration of record- 
ing technics in studios this past year. Compromise has been effected 
between extended high-frequency range and film background noise, 
with the result that an upper cut-off at about 8000 cycles per second 
is generally attempted. The use of microphones of the dynamic type 
continues to expand although the condenser microphone is still used 
for dialog recording in many studios. 

2. New Recording Equipment (Dual System). Several innova- 
tions in the methods of light-valve recording were announced by 
Electrical Research Products, Inc. Changes have been made in the 
input circuit of the light valve so as to compensate for the time shift 
of the effective exposure produced by the light-valve ribbons, which 
is equal to the time required for the film to travel from the neutral 
position of the image of one ribbon. With a constant film speed of 
90 feet per minute and a ribbon spacing of one mil, the loss in effec- 
tive exposure is dependent upon the frequency of the input current, 
amounting to approximately 3 db. at 9000 cycles. Compensation 
for this loss is accomplished by splitting the input circuit to the 
light valve and inserting a delay in the part of the circuit that 
feeds the upper ribbon. The delay is adjusted so that maximum re- 
sponse occurs at 8000 cycles. With this adjustment it so happens 
that the ribbons are approximately 180 degrees out of phase at 9500 
cycles, the frequency at which they are tuned. This out-of-phase 
relation allows a greater input to the light valve at the higher fre- 
quencies with less danger of light-valve clash. 

Amplitude adjustment relative to the frequency is introduced so 
as to utilize the maximum volume range available in the film material 
and in the recording system with phase-adjusted light valve as above 
described. The energy distribution of orchestral music or of speech 
is such that it is feasible to record with a rising characteristic while 
still maintaining constant probability of valve-ribbon clash at all 
frequencies. Since most of the annoyance caused by background 


noise is due to the high-frequency components of the noise, subse- 
quent amplitude adjustment in the reproducing system to com- 
pensate for the amplitude adjustment introduced during recording 
reduces the reproduced noise level and increases the signal-to-noise 
ratio considerably. 

A study of recording conditions attainable with the phase-adjusted 
light valve and amplitude adjustment has resulted in the design of a 
recording system in which the 8000-cycle response is increased about 
28 db. The total harmonic content is reduced about 12 db. relative 
to that in previous systems. 

FIG. 3. Portable recorder showing the film path (RCA 
Victor Co.). 

The use of toe recording up to the point of preparing the master 
negative by re-recording further improves quality by eliminating a 
printing operation and its consequent losses and distortions. 

The RCA Victor Company announces a light-weight dual film 
recording equipment. This recorder combines portability and light 
weight with the merit of the rotary stabilizer type of drive, which 
effects uniform motion of the film at the exposure point. The neces- 
sary controls are built into the base of the recorder (Fig. 3). The 


optical system produces the symmetrical type of sound track ob- 
tained with standard studio equipment. The standard galvanome- 
ter is used. This has integral provision for noise reduction, if a 
small noise reduction amplifier is added to the main amplifier de- 
scribed above. 

A-c., d-c. interlock camera motors equipped with the regular 
cine" type mountings are used. These motors operate on 115 volts 
d-c., derived from a set of three 45-volt .B-batteries. For studio 
operation, 110- volt, d-c. mains may be plugged in. 

To reduce the weight further no tripod is supplied for the recorder, 
and the latter is designed to operate standing on the side or end of 
the carrying case. Provision has been made to fasten a plate to the 
base to act as an adapter to any kind of tripod, if desired by the user. 

3. New Recording Equipment (Single System). During the past 
year the RCA Victor Co. has introduced a portable single-film record- 
ing system, meant chiefly as a portable newsreel recording equipment 
for use with an Akeley audio camera. The equipment is noted for 
its simplicity, ruggedness, and light weight. Inductor type pressure- 
operated microphones are used, and provisions are made so that 
the outputs of two of these microphones may be mixed simulta- 
neously. The microphones are of the permanent-magnet type, com- 
pact, unusually rugged, free from noise due to shock excitation, and 
are little affected by wind. Since it is a relatively high-level, low- 
impedance device, it does not need an amplifier closely associated 
with it, but can utilize a transmission line to the amplifier. 

Because of the points mentioned above, pre-mixing is effected in the 
main amplifier, and no separate microphone amplifiers are used. 
This results in a saving of bulk, battery consumption, and amplifier 
stages. A selector key is provided so that when only one microphone 
is used, the mixer control for the other is automatically eliminated 
from the circuit, thereby increasing the available gain. This system 
provides the maximum signal-to-hiss ratio under all conditions. 

The amplifier itself is quite compact, uses non-microphonic radio- 
trons, and has an output in excess of the normal requirements for a 
standard studio type of galvanometer. The standard studio type of 
galvanometer is employed in the small optical system that fits the 
Akeley audio camera. 

Electrical Research Products, Inc., has announced the Western 
Electric Type G single-film portable recording system, which is 
primarily intended for use in newsreel work where lightness, ease of 


operation, reliability, and high-quality results are the controlling 
factors. The normal sound recording part of the system consists of a 
microphone and tripod, an amplifier assembly, a motor battery case, a 
monitoring head-set, a modulator unit, and the necessary connecting 
cords. This equipment has a total weight of less than one hundred 
pounds, and is relatively convenient to transport as personal bag- 
gage by automobile or train, even when the camera and its tripod and 
the motor battery are added. 

This light-weight recording system was made possible by the 

FIG. 4. Wave analyzer (General Radio Co.). 

development of the moving-coil microphone, which does not require a 
transmitter amplifier, light-weight speech transformers, which are 
used in the compact high-gain amplifier, and the permanent-magnet 
light valve, which is used in the modulator unit. Although this is an 
exceptionally portable and compact recording system, the reliability 
of operation and quality of recording are in no way impaired. 

4. Accessories. The General Radio Company has announced the 
Type 653 Volume Control. These mixer controls have been de- 
veloped in order to reduce to its lowest possible value electrical 
noise introduced into the sound system. They are of the step-by-step 
design, having contact points and switches of approximately the same 



[J. S. M. p. E. 

alloy, so that electrical contact potential is reduced to zero. The 
switch is of the four-blade construction, and is so cut as to provide 
a sidewise wiping contact on the switches to prevent cutting and to 
keep them clear of dirt. The windings are on bakelite posts molded 
into the switch contact structure. This reduces the possibility of 
breaking the resistance wire due to mechanical strain. A ladder cir- 
cuit is used, which has a continuously variable attenuator of about 
I 1 / 2 db. per step over the first thirty steps, the last three steps being 
in larger increments to the cut-off or infinity attenuation position. 

The same company announces the Type 636-A wave analyzer, 
which is a precision instrument to facilitate measurements of har- 
monic distortion in the audio-frequency circuits (Fig. 4.) The 
instrument operates on the heterodyne principle, in which the fre- 

FIG. 5. 

Sound meter and frequency analyzer (Electrical Research Pro- 
ducts, Inc.). 

quency under analysis, which may be anywhere in the range from 5 to 
15,000 cycles, is heterodyned by means of a local oscillator to a 
frequency of 50,000 cycles. The 50,000 cycles is passed through a 
very highly selective amplifier, which is tuned interstage by means of 
quartz crystals. Two quartz crystal tuning stages are used. The 
amplitude of each harmonic present in the voice-wave can be iso- 
lated and measured on the output vacuum-tube voltmeter. Push- 
pull detection is used, and particular care has been exercised in the 
design of the detector so that it, of itself, introduces little if any dis- 
tortion. The selectivity is very high, the discrimination to fre- 
quencies only 2 cycles off resonance being 6 db. At 100 cycles off 
resonance, the discrimination is over 60 db. This means that 60 


cycles and the harmonics thereof can be measured with ease; and 
the tenth and eleventh harmonic, for example, can be separated by 
more than 50 db. The analyzer is entirely self-contained, and re- 
quires no outside equipment of any sort except a 6-volt storage battery 
for filament supply. 

The new Western Electric crystal-controlled sound frequency 
analyzer is shown in Fig. 5. It is essentially a band-pass filter of fixed 
width, which is continuously variable throughout the audible fre- 
quency range: namely, 40 to 10,500 cycles per second. Either of two 
band widths: namely, 20 cycles per second or 200 cycles per second, 
may be selected by operating a pair of keys. The general circuit 
arrangement, while new in application, is not unusual in principle 
except as to the use of quartz crystals to obtain stability and high 
discrimination. In the analyzer unit, the incoming signal is hetero- 
dyned by a variable high-frequency oscillator, and the modulation 
produced passed through the crystal filter. Another portion of the 
tuned oscillator output is used to demodulate the output of the 
crystal filter. The output of the demodulator is fed into the final 
amplifier in audio-frequency form, and may be monitored and 
measured in the sound meter or recorded in a high-speed level 

The excellence of this analyzer is claimed to be due to the ex- 
ceptionally high suppression attained outside the band of frequencies 
passed by the crystal filters. Taking into consideration harmonic 
generation in the tubes and other limiting factors, a suppression of 
50 db. is actually obtained at 45 cycles either side of the center of the 
20-cycle band. The use of this instrument in conjunction with a 
sound meter or high-speed level recorder permits the rapid solution 
of a large number of different types of problems, such as the analysis 
of the sound spectra of various types of noise or of musical instru- 
ments, the location of resonance effects in auditoriums, and the 
measurement of harmonics in sound systems. 

The high-speed level recorder referred. to above is a development 
of the Bell Telephone Laboratories. It will record rapid changes in 
audio-frequency currents directly on a moving strip of waxed paper by 
means of a stylus. A 60-db. range of intensity can be covered, during 
which range the deflection of the stylus is proportional to the input in 
decibels. It will follow changes of intensity at adjustable rates 
up to 360 db. per second. Various fixed paper speeds may be selected. 

The equipment consists of three units: namely, an amplifier, a 


recorder, and a power-supply unit, each provided with an aluminum 
carrying case for field use or arranged for rack mounting for laboratory 

This instrument is extremely useful for a wide variety of acoustical 
measurements. It has all the advantages of any direct recording meter 
in that there is no time-consuming developing process as with photo- 
graphic recorders. The record is instantly available and observable 
during the recording. A partial list of its uses includes : reverbera- 
tion time measurements and studies of the pattern of sound decay in 
auditoriums under various conditions; high-speed recording in 
conjunction with sound frequency analyzers; noise measurement 
where the records of changing conditions of noise are desired ; studies 
of the intensity of singing and speaking voices; and loud speaker 
calibration when synchronized with a variable-frequency oscillator. 

A device has been made available by Electrical Research Products, 
Inc., for automatically altering the volume or frequency spectrum of 
the signal in a particular circuit under the control of a signal in an- 
other part of the circuit. A typical application of this function is 
found when two or more sound records are re-recorded and combined, 
or when original sound is recorded in combination with sound al- 
ready recorded. One particular application, for instance, would 
be in the case of a picture showing a couple dancing to a fairly loud 
orchestra and conversing in an ordinary tone of voice. As normally 
recorded, the speech under such conditions would be practically 
unintelligible if the music were mixed at a satisfactory uncontrolled 
volume. However, if the background sound of the orchestra can be 
reduced appreciably in the presence of the speech, the latter be- 
comes easily intelligible and a very satisfactory effect of loud back- 
ground music is obtained. It is obviously impossible for the mixer 
operator manually to control the musical background to produce the 
effect satisfactorily. By means of the voice-operated switching 
device, the musical background is suppressed only during the speak- 
ing. Under these conditions the volume changes of the music are 
scarcely noticeable and the over-all effect is as required. 

B. Amateur 

The RCA Victor Company has announced a new 16-mm. sound 
camera and associated recording equipment. There are two types of 
this camera : namely, the A utophone Type, by which only the speech 
of the operator can be recorded, and the Microphone Type, by which 

June, 1934] 



the sound of the photographed subject may be recorded. Two speeds 
of operation, 24 or 16 frames per second, are provided. In addition 
to the camera the following equipment is supplied with the Micro- 
phone Type: a two-stage amplifier with cable, a microphone with 
cable, a battery box, a belt assembly, a connecting cable from camera 
to amplifier, and one monitoring phone. 

The amplifier is a high-gain, battery-operated type using one RCA 
232 and one RCA 233 tube. Special features are its manual volume 
control and visual recording level indicator, the latter consisting of 
three neon tubes, each lighting at a different sound level. 

FIG. 6. British 16-mm. sound-on-film camera (Movies and House Talkies, 


The first announcement of a British 16-mm. sound-on-film camera 
appeared in the April, 1934, issue of Home Movies and Home Talkies. 
Variable-density recording is used, the recording lamp being quickly 
removable for shipment or replacement. A four-lens turret is 
fitted and direct focusing is employed. Film retorts of 400-ft. ca- 
pacity are supplied. Single-perforation S. M. P. E. standard film 
is utilized 35 * (Fig. 6). 


1. New Sound Equipment. Little has been offered in the way of 
new sound equipment by the manufacturers in the last year. The 



[J. S. M. P. E. 

high-fidelity reproducing system is being installed in an increasingly 
large number of theaters, and up to March 3, 1934, 717 theaters had 
been equipped with wide-range projection equipment. 

Of interest to studios and theaters showing double-film previews 
is an announcement of a new type of double-film attachment that 
permits running the picture and the sound print in synchronism over 
one projection machine (Fig. 7) . 

FIG. 7. Double-film attachment for pre- 
view projection (Electrical Research Prod- 
ucts, Inc.}. 

2. New Projectors and Accessories. From Germany comes the 
announcement that the firm of Zeiss Ikon produced in 1933 a new type 
of theater projector called Ernemann V, its special feature being a 
water circulating scheme for preventing the film gate from becoming 
hot, as well as certain arrangements which, in case of the advent of 
wide film, would make it possible to adapt the 35-mm. film track for 
films of larger sizes. The water flows mainly through such parts as 
are heated by the projection lamps, especially the film track and gate, 
as well as the parts to be oiled and the transport mechanism. The 
projector is provided with a rotary shutter between the light and the 
film. Descriptions of Ernemann V and Ernon L V will be found in 


Zeiss Ikon have also improved their projection lamps and their well- 
known Artisol lamps with the carbon points set at an obtuse angle, 
and a reflecting mirror with condenser. This involves a special 
correction of the reflecting mirror which permits very uniform illumi- 
nation of the gate and does not require an additional condenser. 
The Kinesol lamp can be used with a maximum current of 35 amperes, 
and the Artisol lamps have recently been delivered for amperages up 
to 80. 

During the past year, the firm of Eugene Bauer, Stuttgart, Ger- 
many, which was amalgamated with the Kinoton Company in Ger- 
many, has produced three new projectors, two of which have not yet 
been completely described in the literature. 

The three new projector models are called Standard 5, Standard 7, 
and Super 7. The Standard 7 is described in Kinotechnik* 1 It 
might be mentioned that Standard 5 is intended for use in medium- 
sized theaters, and is provided with a rotary shutter between the film 
and the lamp, as well as a completely enclosed casing with an auto- 
matic circulating lubricator. It can be used with a lens of 62.5-mm. 
focus. In view of its particular suitability for smaller cinemas, this 
apparatus is provided with take-up spools for 1300 meters of film, 
thus enabling a long uninterrupted projection. Bauer points out 
that an important novelty of this projector is the framing device, 
according to which the film gate moves together with the Geneva 
cross, whereas the picture gate remains in a fixed position. This 
arrangement permits framing even if the projector is not in operation. 

An important feature of the Standard 7 projector is the incorporated 
sound pick-up, which is rigidly connected to the projector. The 
construction of the sound pick-up is similar to that of the LT type 
put out by the firm of Bauer. The Standard 7 projector, however, can 
be used also in conjunction with any other sound pick-up without 
having to make alterations. It is driven by means of a shaft that 
makes 1440 rpm. on practically the same principle as that of the 
Standard 5 projector. 

The Super 7 projector, like the Standard 7, is suitable for use in the 
largest theaters, and a particularly interesting feature of the former 
is the completely enclosed film track. The projector can be fitted 
with lenses up to 104-mm. focus, and is equipped with a new lamp 
casing which is large enough even for lamps with a mirror of 300 mm. 

The firm of Bauer has also improved its Bauer-Kohlennachschub 
N 2. A special wiring system in this arc lamp provides an auto- 


matic control of the carbons. Other mirror lamps are described in 
Filmtechnik especially one produced by the firm of Erko, which em- 
ploys a mirror of 250-mm. focal length constructed by the firm of 
Busch, the carbon points of which are axially arranged. 

In the United States, C. Tuttle has published in the JOURNAL 40 
an analysis of distortion in theater projection, concluding with the 
opinion that the amount of distortion tolerable from the view- 
point of the observer is greater than ordinarily supposed. Distortion 
from the cameraman's view-point, with particular reference to the 
keystone effect in theaters, is the subject of a paper by R. F. 
Mitchell. 41 

Two new projector arcs for the new a-c. carbon arrangement have 
been announced, and both are equipped with elliptical reflectors. 42 

FIG. 8. Sound projector for 16-mm. film (Bell & Howell Co.) 

A water-cooled metal reflector, for which claims of unusual durability 
are made, has been developed in England. 43 Some work leading 
toward the development of new reflecting surfaces on glass, par- 
ticularly of aluminum and magnesium, has also been reported during 
the past year. 44 

Another process of stereoscopic projection, invented by G. Jellinek, 
has been described in the literature. 45 C. R. Haupt has also pub- 
lished a comprehensive discussion of the question. 46 The usual 
number of patents on stereoscopic projection have been noted. 47 
There have been a few patents also on non-intermittent projection. 48 

Two articles have been published on optical systems for trans- 
parency projection in the studio, 49 which suggest means for lessening 


the "hot-spot" effect on the translucent screen used as the back- 
ground. Opinions seem to agree that no cure has yet been found 
for this defect, and the subject appears to need further investigation. 
3. 16- Mm. Sound-on- Film Projection. The adoption of 16-mm. 
sound-on-film by the public appears to be taking place gradually, 
and new equipment continues to be introduced every few months. 
The Bell & Howell Company has announced the Filmosound pro- 
jector (Fig. 8). This is a 16-mm. projector and sound reproducer, 
embodying a 500- or a 750- watt lamp, a sound reproducing head, and 

FIG. 9. Disk reproducer for 16-mm. film projector (Western Elec- 
tric Co. of London). 

amplifier system in one case; and loud speaker in a separate case, 
which can also house all cables necessary for operation, as well as 
spare accessories. Features of the apparatus are an "optical slit," 
which is obtained through a system of cylindrical lenses so arranged 
as to produce an image of the exciter filament reduced in size approxi- 
mately 10 to 1 in the vertical plane, and an oscillatory circuit supply- 
ing a high-frequency alternating current for the exciter lamp and 
incorporating a single 145 triad to eliminate hum due to a-c. supply. 
The volume is controlled by simultaneously vary ing the photo-cell and 
exciter lamp voltages with a single control. 

From England we are advised that the Western Electric Company, 



[J. S. M. P. E. 

Ltd., is manufacturing 16-mm. sound-on-disk apparatus (Fig. 9). 
Some of the features of the equipment are : a double projector system 
for change-over purposes, thus allowing a continuous program to be 
given; a gear-changing device to enable the projector to run at either 
16 or 24 frames per second; and a gear-changing device to allow the 
turntable to run at either 33 Vs rpm. for sound pictures, or 78 rpm. for 
incidental music. 

A sound-on-film projector for 17.5-mm. film and using one square 
perforation per picture was introduced abroad last winter by the 
Pat he Company. A gate mask is used to prevent the perforation 
holes from showing on the screen. One of the most interesting 

FIG. 10. Pathe 17.5-mm. sound-on-film projector (Movie and Home Talkies, 


features is that the same lamp is used both for projection and illumina- 
tion of the sound track. After the light passes through the sound 
track, it is reflected by a mirror into the photo-cell in the sound-head. 
Volume control is secured by a rotatable shutter in the light path 
before the photo-cell 50 (Fig. 10). 

Early in 1934, the German firm of E. Bauer introduced a port- 
able 16-mm. sound-on-film projector. The sound-head permits 
reproduction to 6000 cycles. The projector is operated with an 
Asynchron motor and the apparatus weighs about 60 pounds. 


1. Film Development. There is little new to report as to the 
adoption of new practices related to the development of 35-mm. 

June, 1934] 



sound or picture film. Some properties of two-bath developers for 
film have been discussed in the JOURNAL. 51 Three types of two-bath 
developers have been investigated as follows: (7) bath A containing 
all the developing agents plus sodium sulfite; bath B, all the alkali 
plus the balance of the sulfite ; (2) both baths containing developing 
agents; (3) both baths of identical composition, the first bath 
being ^replaced by the second as it becomes exhausted. The results 
of the investigation showed that the type (1) is the most satisfactory 
developer combination, and with this method it is possible to obtain 
an almost constant gamma with only a slight loss of emulsion speed 

FIG. 11. 

Densitometer employing photronic cell (Para- 
mount Productions). 

over a fairly wide range of time of development. A formula is also 
suggested for the development of variable density sound negatives. 
The application of two-bath developers to machine and rack-and-tank 
systems is described. 

2. Laboratory Equipment. There is little to report in the way of 
new laboratory equipment during the past year. From the Para- 
mount Laboratory in Hollywood comes an announcement of a new 
type of visual densitometer (Fig. 11). Its essential elements con- 
sist of a light source, a Weston Photronic cell, and a microarrimeter. 
The density range of the densitometer is to 1.00 density diffuse. 
It is claimed to be extremely accurate and stable, and very simple to 


operate. One major advantage over the eye-balance system is 
that it obviates all personal error caused by eye fatigue. 

Extensive studies of printing problems have been made at the Bell 
Telephone Laboratories during the past year, reports of which have 
been published in the JOURNAL. 52 It has been definitely established 
by this investigation that deterioration in sound printing in standard 
printers can be attributed to improper sprocket hole dimensions of the 
negative and positive films, which are in contact. The report showed 
that a certain fixed differential in pitch amounting to 3.6 thousandths 
of an inch should exist between the negative and the print stock if 
serious distortion of the sound wave envelope is to be avoided. The 
negative film being adjacent to the curved sprocket wheel should, of 
course, be of lesser pitch than the positive which is on the outside with 
a larger radius of curvature. 

These studies have created a new interest in film pitch dimensions 
in the studios and film laboratories. The possibility of introducing a 
negative sound film having a sub-standard pitch of 0.1866 inch as 
compared with the standard value of 0.1870 inch, has been presented, 
and this may be a solution of the problem of obtaining correct sprocket 
hole pitch relations between negative and positive films. 


1. Education. A sound film project initiated by the University 
of Chicago in 1932 for the purpose of presenting the subject of natural 
science to the entire student body by means of lectures only was re- 
ported to be successful in a paper by Lemon read last fall at the 
Chicago meeting. Plans were being made to extend the work to 
cover social, biological, and physical sciences. 53 Two interesting 
books on the subject of the use of sound pictures in education are 
listed in the Bibliography section of this report. 

At the Century of Progress Exposition in Chicago in 1933, motion 
pictures were reported to have been used effectively in a great 
many of the concessions and educational exhibits. 54 

2. Timing Devices. The Western Electric timing system men- 
tioned in last year's report has been further developed through co- 
operation of the Eastman Kodak Company and the Bell Telephone 
Laboratories (Fig. 12). It is primarily intended for use in ac- 
curately timing any sequence of events that can be photographed. 
Examples of particular applications are: the timing of foot 
races, airplane races, and other sporting events; the time analysis of 


movements made in playing games or in performing various kinds 
of work; industrial applications, such as the study of processes and 
reactions in such fields as physics, chemistry, biology, and psychology; 

FIG. 12. Timing system, attached to Eastman high-speed cam- 
era (Western Electric Co. and Eastman Kodak Co.}. 

FIG. 13. Records made with Eastman high-speed camera and Western 
Electric timer: ignition of Mazda photoflash lamp connected to electric cir- 
cuit, which in turn ignites successively two other lamps with their glass 
surfaces in contact. Total time interval 0.16 sec. (Western Electric Co.). 

as an acceleration microscope to obtain data for plotting time against 
displacement or deflection; and various specialized applications in 
widely varied lines. Fig. 13 shows the time lapse occurring during the 

370 PROGRESS COMMITTEE [j. s. M. p. E. 

ignition of three photoflash lamps, one of which was fired electrically. 

The system consists fundamentally of two main elements: namely, 
a special 16-mm., high-speed, non-intermittent motion picture camera 
and an electrical equipment for registering time. The camera has two 
lens systems, so arranged that when the camera is used to photograph 
any event, the moving dials of the time register are also simul- 
taneously photographed on one-eighth of each frame of film beside 
the picture. Upon viewing or projecting the film so made, the in- 
stant at which any portion of the event occurred can thus be read 
immediately from the direct time record in the picture. 

The camera is available in two models: namely, a high-speed model 
known as Type No. 1, taking from 30 to 250 frames per second, and 
an ultra high-speed model, known as Type No. 2, taking from 300 to 
2000 frames per second. 

For athletic events, or other events occupying time intervals not 
exceeding five minutes, the accuracy of the timing system justifies 
reading the time register to the nearest scale division: namely, 
Vioo of a second. For longer intervals, and provided it is possible 
to operate the system in locations where the temperature is between 
40 and 90F., the frequency generator which supplies power to the 
time register is accurate to 1 part in 100,000. For engineering use a 
time register is available with dials graduated in seconds and l / m of 
a second, for use with the ultra high-speed No. 2 camera. When us- 
ing this time register and camera, it is feasible, by interpolation, to 
determine elapsed time to an accuracy of Vaooo of a second. If 
desired, the time register can be used alone without the camera, after 
the manner of using a stop-watch, or with two or more cameras 
operating in synchronism. 

3. Miscellaneous. An interesting development is reported in 
Editor & Publisher. A method has been developed that permits a 
bit more than eight full-sized newspaper pages to be recorded on a 
strip of film measuring ! 3 /s by 12 inches. Despite the extreme reduc- 
tion of the page size, the clarity of the image on the film is so great 
that by means of a viewing device any part of the original page can 
be projected to 150 per cent of its original size. It is claimed that 
all the pages of a month's file of a 50-page newspaper could be re- 
corded on a film that would occupy storage space 3 5 /s by 3 5 /s by l l / 2 
inches. A photograph of the viewing device for examination of the 
35-mm. film image of the newspaper library is shown in Fig. 14. 
One month's file is shown on the small reel. 

June, 1934] 




Possible duplication of published information will be avoided 
in future as a result of the joining of the Motion Picture Projectionist 
(New York) with the International Projectionist (New York). This 
trend is a favorable one and was begun several years ago when the 
Motion Picture World combined with the Exhibitors Herald. A 
new amateur publication, Personal Movies (Canton, Ohio), which 
was issued first in 1932, has been continued. Those who desire to do 

FIG. 14. Viewing device for 
examining 35-mm. film image 
of newspaper library (East- 
man Kodak Co.}. 

reference reading will welcome the news that a ten-year index ap- 
peared in 1933 for Vols. 1 to 10 of the Royal Photographic Society's 
Journal, Photographic Abstracts. A list ot the principal books that 
have been published since the last report of the Committee (April, 
1933) follows: 

1. Year Book of Motion Pictures 1934, 16th Edition ; Film Daily, 
New York. 

372 PROGRESS COMMITTEE [j. s. M. p. E. 

2. Motion Picture Almanac 1933; Quigley Publishing Co., New 

3. Kinematograph Year Book 1934; Kinematograph Publications, 
Ltd., London. 

4. Yearbook of Photography, Cinematography, and Reproduction 
Processes for the years 1928-1929 (Jahrbuch fur Photographic, Kine- 
matographie, und Reproduktionsverfahren fur die Jahre 1928-1929), 
Vol. 31, Pt. 2, edited by J. M. Eder, E. Kuchinka, and C. Emmer- 
mann; W. Knappe (Halle). 

5. Yearbook of the Cine-Amateur 1934 (Jahrbuch des Kino- 
Amateur 1934) , edited by W. Frerk; Photokino-Verlag, Berlin. 

6. The Film and Its World (Der Film und seine Welt), German film 
Almanac, edited by F. Henseleit; Photokino-Verlag, Berlin. 

7. Studies from the Emulsion and Colloid Laboratory (in Russian) 
Vol. 1, 1932; Kinophoto Institute, Moscow. 

8. Pictorial Composition in Photography, by A. Hammond; Ameri- 
can Photographic Publishing Co., Boston, Mass. 

9. Commercial Cinematography, by G. H. Sewell; Pitman & Sons, 

10. Sound Picture Recording and Projection, by K. M. Macllvain; 
International Text Book Co., Scranton, Pa. 

1 1 . Talking Pictures, 2nd Edition, by B. Brown ; Pitman & Sons, 

12. Illumination for Motion Picture Projection (La Lumiere dans 
la Projection Cinematographique), by J. Marrette; Gauthier-Villars, 

13. Motion Picture Projection and Sound Pictures, by J. R. Cam- 
eron and others; Cameron Publishing Co., Woodmont, Conn. 

14. Theatre and Motion Pictures, several articles by various au- 
thors; Encyclopedia Britannica, New York, N. Y. 

15. The Visual Fatigue of Motion Pictures, by A. E. Singer; 
Amusement Age Publishing Co., New York. 

16. Amateur Talking Pictures and Recording, by B. Brown; I. 
Pitman & Sons, London. 

17. Movie Making Made Easy, by W. J. Shannon; Moorfield & 
Shannon, Nutley, N. J. 

18. The Sound Motion Picture in Science Teaching, by A. J. 
Rulon; Harvard University Press, Cambridge, Mass. 

19. The Educational Talking Picture, by F. L. Devereaux; Univ. 
of Chicago Press, Chicago, 111. 

June, 1934] PROGRESS COMMITTEE -373 

20. Film Tricks and Trick Films (Filmtricks und Trickfilme), by 
A. Stuler; W. Knapp (Halle). 

21. Infrared Photography, by S. O. Rawling; Blackie & Sons, 

22. The Photography of Colored Objects, 12th Edition, Revised; 
Eastman Kodak Co., Rochester, N. Y. 

23. The Complete Projectionist, by R. H. Cricks; Kinemato- 
graph Publications, Ltd., London. 

24. The Projectionist's Handbook, by R. Pitchford and F. Coombs; 
Kinematograph Publications, Ltd., London. 


A . General Field of Progress of the Motion Picture Industry in 
Great Britain 

Resume. The year 1933 has been a successful one, in general, for 
the British industry. On the production side the technical standard 
has improved considerably, resulting in pictures of a better class 
than have previously been made in this country. In the theater field 
conditions have generally improved, but some of the smaller exhibitors 
experienced trying times owing to the fact that so many large cinemas 
are now being built. The year has shown a marked increase in the 
number of British films shown in cinemas in this country. Of the 
685 pictures shown to the trade, 456 were of American origin, 196 
British, and 33 from other parts of the world. 

16- Mm. Development. The principal technical activities in this 
field have been in connection with the development of sub-standard 
equipment. Several such equipments have appeared on the market 
employing both sound-on-disk and sound-on-film in the 16-mm. size. 
Considerable lack of standardization is at present evident with regard 
to the latter. Equipments have been designed so as to be adaptable 
to a considerable range of voltage, and to operate at both silent and 
talking picture speeds. 

The use of films for advertising has developed rapidly along the 
lines of improved technic of production and the production of films 
suitable for non-theatrical exhibition, by means of sub-standard 
equipment or daylight trucks. The principal enterprise in this direc- 
tion has been shown by the tobacco companies, automobile manu- 
facturers, and manufacturing confectioners. Little progress has 
been made in connection with the educational use of films on account 
of lack of Government funds. 


Studio Production. The progress in the studios was greater during 
1933 than at any time in the history of motion pictures in this country, 
and 33 more feature productions were registered than in 1932. 

The standard of technic definitely improved and productions be- 
came more polished than previously, as shown by the success in 
America of some of the features made here. Undoubtedly the ar- 
rangement between a leading American company and some producing 
companies over here, whereby the American company guarantees a 
release in the United States for first-class British productions, appears 
to be partly responsible for the producers' attempts to make films of 
high quality. The leading companies are spending from 50,000 to 
80,000 on a production, in order to produce material comparable 
with American feature productions, while fewer shorts are being 

Technically, there has been considerable improvement in studio 
products due in part to improved lighting equipment, better use of 
incandescent lighting, more expert photography, general improvement 
in sound, more elaborate and better designed sets, and improved 
laboratory work. 

Studio Expansion. The majority of studios are working at full 
capacity, and some are increasing their facilities in order to fulfill the 
demands for studio space. Two new studios were built, one at 
Hammersmith for P.D.C. having two stages, and the other at Shep- 
perton, while other studios are considering new buildings, and one is 
being converted for sound. 

Most of the larger studios have now installed the latest type ol 
camera cranes ; and a new camera which is giving excellent results 
has been put on the market by the British firm of Messrs. Vinten, 

Back projection is now commonly used, although a satislactory 
medium to take the place of the glass screen has not yet been found. 
Oiled paper, chemically treated linen, and cotton fabrics have been 
tried without much success. 

Theater Progress. In the main, 1933 has been a poor year for ex- 
hibitors, for, although business conditions generally have improved, 
this has been offset by the fact that during the year the weather was 
extraordinarily good, as a result of which people preferred other forms 
of entertainment. The smaller cinemas, in particular, are finding 
conditions difficult, one reason for which is the entertainment tax, 
which imposes a burden they find difficult to carry. In spite of this 


the total seating capacity in this country is about the same, for fewer 
but larger cinemas have been built. Some 75 new theaters were 
erected, and the number in the course of construction at the moment is 
83, with an average seating capacity of 1335. 

There is little or no indication that the general public is tiring 
of talking pictures, but they are showing considerable discrimination 
in choosing films they wish to see and, because of the luxurious cine- 
mas now being built, are becoming educated to a degree of comfort 
that hitherto they have not expected. 

One of the problems that has to be met is that of over-seating; 
for, with the growth of the suburbs surrounding the larger towns, it is 
quite a common occurrence to find two or three cinemas with fairly 
large seating capacities catering to a population that is not large 
enough to patronize all the attractions offered. The Cinematograph 
Exhibitors' Association is concerned with this problem, and the 
general feeling is that the more modern theaters will gradually cause 
the small, independent, and out-of-date theaters to disappear. 

B. Progress in the Motion Picture Industry in Japan 

In looking at the activities and progress of the motion picture 
industry in Japan one must consider the local market for motion 
pictures. Simply to compare the progress made in production in 
Japan with that in the United States does not give the complete 
story. The price of admission to a "movie" in the cities of Japan 
begins at five sen and goes to thirty or forty sen; in the smaller towns 
five and ten sen are the usual prices. The average number of release 
prints is seven to ten to a picture. All this means that the cost of 
the average first-class picture must not exceed 30,000 or 35,000 yen. 
All cameras, printers, film, and even developing agents, must be 
imported and paid for in pounds, francs, marks, or dollars, all of 
which come high in yen. The waste in film can not be very high: 
50 per cent of the negative footage is the average in first-class studios. 
This leaves very little room for retakes or attempts to improve by 
experiment. Nor does it leave much money for development work 
or for buying the latest equipment. 

Thus, when sound came along it was hardly feasible for Japanese 
producers to buy equipment priced for United States consumption 
and with royalty payments required that were suited to a 200-400 
print release per picture. As a result, development of sound in 
Japan was delayed, and attempts were made to develop sound ap- 

376 PROGRESS COMMITTEE [j. s. M. p. E. 

par at us locally. Developments were naturally rather slow, 
was not until 1932 that any of this apparatus was in condition to be 
exploited. During 1932, Shochiku Studios produced several pictures 
in sound on equipment developed and assembled in their studios by 
the Tsuchichashi Bros. Also by 1932, both RCA Victor and Western 
Electric had brought in recording equipment. In 1932, Western 
Electric turned out one test picture. 

In local parlance a sound picture is one with scored music or sound 
effects, whereas a "talkie" is a picture to which the sound is syn- 
chronized. During 1933 Shochiku produced "sound pictures" and 
"talkies" regularly in their two studios at Kamata and Kyoto, al- 
though the big percentage of their releases were still silent. Also, 
Western Electric signed up Nikkatsu as a licensee during 1933, 
and they proceeded to turn out one "talkie" a month, which is 
equivalent to one-fourth their total production. One or two equip- 
ments were bought from independent dealers in sound equipment 
in the United States, and several pictures were turned out on these. 
Besides, there have been a number of locally assembled equipments 
used for a few pictures by smaller independent studios. 

There have been several sound stages built during 1932 and 1933, 
and more are under construction. The talkies are drawing the 
biggest share of the patronage and the industry will doubtless be 
forced to turn "all talkie" during 1934. At the end of 1933 a studio 
devoted to trick effects was established in Kyoto. This is the first 
of its kind. In the studios some attempts at process work have been 
made, but there has been little success to date, chiefly for lack of 
equipment. Two companies have been formed for the production 
of animated cartoons but there have not been many releases. 

By the first of 1933 there were two independent laboratories 
equipped with continuous developing machinery and apparatus 
necessary for handling sound negative and positive. One of these 
has a sound stage, as well, and has produced several pictures. The 
bulk of the film, both negative and positive, is developed by rack and 
tank. During 1933, 431 pictures were made and released in Japan, 
of which 159 were sound pictures, including several shorts and news- 
reels. The end of 1934 will undoubtedly see the percentage of sound 
pictures much higher. 


1 Internal. Phot., 5 (Aug., 1933), p. 31. 

2 Phot. Ind., 31 (March 8, 1933), p. 254; Ibid., 31 (June 7, 1933). 


Phot. Korr., 69 (March, 1933), p. 41. 
Set. Ind. Phot., 4 (June, 1933), p. 177. 

* Bur. Stand. J. Research, 10 (Feb., 1933), p. 211; Ibid., 11 (Dec., 1933), p. 743. 
6 Kinophoto Inst., 1 (1932), p. 70; Ibid., 1 (1932), p. 128; Ibid. ,2 (1933), p. 127. 
Also Trans. Opt. Inst., Leningrad, 9 (1933), No. 88, p. 3. 

7 U. S. Pats. 1,884,035; 1,923,495; Brit. Pat. 396,646. 

8 U. S. Pat. 1,912,758. 

9 U. S, Pat. 1,897,838. 

10 U. S. Pat. 1,897,878; Phot. Ind., 30 (Sept. 21, 1932), p. 955. 

"U.S.Pat. 1,883,559; 1,883,562. 

"Brit. Pat. 378,394; Ger. Pat. 565,378; Fr. Pat. 721,911. 

13 Am. Cinem., 13 (Sept., 1933), p. 174. 

" Am. Cinem., 14 (May, 1933), p. 835. 

15 Int. Phot., 5 (May, 1933), p. 20. 

" Brit. J. Phot., 80 (March 10, 1933), p. 125. 

17 Amer. Cinem., 13 (April, 1933), p. 18. 

18 Film fur Alle, 7 (March, 1933), p. 73. 

19 U. S. Pats. Brit. Pats. 

1,882,530 359,474 

1,888,156 376,044 

1,892,162 377,036 

1,897,262 377,537 










20 Kinotechnik, 15 (March, 1933), p. 75. 

21 Kinotechnik, 15 (Oct., 1933), p. 307. 

22 Amer. Cinem., 13 (Sept., 1933), p. 166: Capt. F. M. Williams. 
Phot. Ind., 31 (Sept., 1933), p. 166: W. Dieterle. 
Photofreund, 12 (May 5, 1932), p. 161 : A. Buchholz. 

Phot. Korr., 68 (Nov., 1932), p. 198: J. Plotnikow. 
Brit. J. Phot., 80 (Jan. 13, 1933), p. 15: B. Alfieri. 
22 * Brit. J. Phot., 80 (Oct. 13, 1933), p. 611. 

23 Amer. Cinem., 13 (Oct., Nov., and Dec., 1933). 

24 Amer. Annual Phot. 47 (1933), p. 242. 

25 /. Soc. Mot. Pict. Eng., XVIII (May, 1932), No. 5, p. 655. 

26 Inter. Phot., 5 (May-Dec., 1933). 

27 Phys. Rev., (March, 1933). 

28 Rev. Sci. Instr., 4 (Aug., 1933), No. 8, p. 449. 
280 Kinemat. Weekly, 199 (Sept. 21, 1933), p. 3. 
286 Phot. Ind., 31 (1933), p. 926. 

29 Ger. Pat. 559,029. 


30 U. S. Pat. 1,918,623. 

Fr. Pats. 708,330; 739,461; 738,353. 

Ger. Pat. 558,365. 

Brit. Pat. 393,901. 

32 Kinotechnik, 14 (1933), p. 237. 

33 Bull. Soc. Franc. Phot. Cinem., 3, p. 20; 75 (July, 1933), p. 151. 

34 Kinotechnik, 15 (1933), p. 85. 

35 Kinotechnik, 15 (1933), p. 348. 

350 Brit. J. Color Supp., 28 (March 2, 1934), p. 12. 

366 Home Movies and Home Talkies, 2 (April, 1934), p. 433. 

36 Filmtechnik, (1933), pp. 127 and 240. 

37 Kinotechnik, 15 (1933), p. 34. 

38 Kinotechnik, 15 (1933), p. 333. 

39 Filmtechnik, (1933), p. 187; p. 241; p. 251; p. 304. 

40 /. Soc. Mot. Pict. Eng., XXI (Sept., 1933), No. 3, p. 198. 

41 Amer. Cinem., 13 (Jan., 1933), p. 45. 

42 Internal. Proj., 6 (Dec., 1933), No. 3, pp. 12 and 24. 

43 Kinemat. Weekly, 194 (April 9, 1933), p. 42. 

44 Phys. Rev., 43 (Feb., 1933), p. 205. 

45 Kinotechnik, 15 (Feb. 15, 1933), p. 45; 15 (March 20, 1933), p. 106. 

46 Internal. Phot., 5 (May, 1933), p. 4. 

47 U. S. Pats. 

1,882,424 1,905,469 

1,882,646 1,905,716 

1,882,648 1,916,320 

1.883.290 1,917,246 

1.883.291 1,919,115 
1,899,139 1,927,925 

Brit. Pat. 376,421, 

48 U. S. Pats. 1,918,788; 1,928,255; Brit. Pat. 346,851. 

49 Amer. Cinem., 14 (Jan., 1934), p. 353; Ibid., 14 (Aug., 1933), p. 134. 

50 Home Movies and Home Talkies, 2 (April, 1934), p. 423. 

51 /. Soc. Mot. Pict. Eng., XXI (July, 1933), No. 1, p. 21. 

62 7. Soc. Mot. Pict. wg., XXI (Oct., 1933), No. 4, p. 294. 

63 /. Soc. Mot. Pict. Eng., XXII (Jan., 1934), No. 1, p. 62. 
4 Film Topics, 9 (1933), No. 2, p. 1. 


There is much current activity within the industry looking toward 
the establishment of a new standard of reel length for multiple reel 
subjects (commonly termed feature pictures). Any change in reel 
length would directly affect three important branches of the in- 
dustry production, exchange practice, and, in particular, projec- 
tion and it was with this thought in mind that the Projection Prac- 
tice Committee investigated thoroughly the proposed changes. 

The standard that has been most favorably regarded to date out- 
side the projection field would establish a reel length of 1700 feet 
maximum, and adopt as standard a reel length of 13V2 inches outer 
diameter. The reasons advanced in support of this proposal are: 

(1) It would eliminate the practice of "doubling," or joining present single 
reels, which in many instances involves extensive cutting both before and after 
projection, resulting in a serious loss of footage and in print mutilation; 

(2) Fewer reels would be required for each feature, resulting in: (a) a reduc- 
tion in the number of film leaders and tail pieces, (&) fewer change-overs, thus as- 
suring a smoother show and less film damage, which is greatest at the beginnings 
and endings of reels. 

In the opinion of this Committee the increase of reel footage to a 
maximum of 1700 feet (which probably would result in a minimum 
footage of 1500 feet, or less) is unsuitable for the general needs of 
the industry and undesirable from the standpoint of the practical 
projectionist. It is quite apparent, as will be demonstrated subse- 
quently, that this proposal decidedly would neither eliminate the 
practice of "doubling," nor offer advantages of sufficient import to 
justify a change in the present reel length standard. 

The outstanding claimed advantage of a 1700-foot reel length is the 
elimination of "doubling." Investigation by the Projection Practice 
Committee discloses the fact that no such result would ensue, for the 
following reason: 

The standard projector magazine, used in all theaters, is 18 inches 
in diameter and has a film-carrying capacity of 3450 feet. The pro- 
posed new reel standard is based on a maximum length of 1700 feet 

* Presented at the Spring, 1934, Meeting at Atlantic City, N. J. 




of film, with the minimum length likely to approximate 1500 feet, or 
less. Obviously, even if the great majority of reel lengths should 
attain the maximum length of 1700 feet and it is very unlikely that 
they would it still would be possible to "double" two such reels and 
remain within the capacity of present standard magazines. Thus is 
vacated the major advantage claimed of the proposed new standard. 

The Projection Practice Committee believes that if it be found de- 
sirable to establish a new standard of reel length, the maximum length 
should be fixed at 2000 feet, with the minimum length in no instance 
less than 1750 feet. Obviously, it would be impossible to "double" 
any two reels that conformed to this recommendation. 

Another advantage claimed for the 1700-foot reel is the reduction 
in the number of reels necessary for mounting and shipping feature 
pictures, and in the number of change-overs required for projection 
the latter reason being advanced in the interest of a smoother show 
than at present. Here again the weight of evidence favors the longer 
reel length, as may be seen from the following comparative table : 


Comparison between Number of Reels for Features of Various Lengths, for Reel 
Lengths of 1700 and 2000 Feet. 

Feature Length 

Cutting Division 

No. of Reels 

6000 Feet 

3 X 1700 plus 1 X 900 
3 X 2000 


6500 Feet 

3 X 1700 plus 1 X 1400 
3 X 1800 plus 1 X 1100 



7000 Feet 

4 X 1500 plus 1 X 1000 
3 X 1900 plus 1 X 1300 



8000 Feet 

5 X 1600 
4 X 2000 



9000 Feet 

5 X 1600 plus 1 X 1000 
4 X 1900 plus 1 X 1400 


12,000 Feet 

7 X 1600 plus 1 X 800 
6 X 2000 


The conclusions to be drawn from Table I, in every instance repre- 
senting the most advantageous basis for each standard, are as follows : 

(a) In no case are more 2000-foot reels required than 1700-foot reels. 
(6) In the majority of cases fewer 2000-foot reels are required than 1700- 
foot reels. 

(c) Flexibility in cutting point is pronounced in favor of 2000-foot reels. 


(d ) Advantages of the 2000-foot reels on longer features are particularly out- 

The following additional reasons are offered by this Committee in 
support of its recommendation for the longer, or 2000-foot, length, 
as against the 1700-foot length: 

(L) Present standard reels, retained by the theater, are 15 inches in diameter 
and will accommodate 2300 feet of film. 

{2) Present standard rewinders will accommodate the 15-inch reel. 

(3) Present standard projector magazines and take-ups will require no change. 

(4) None of these equipment items will occasion any additional expense to 

It is apparent that, over-all, the recommendations of this Committee 
in favor of the longer reel length would: (1) be advantageous in 
shipping and handling film, (2) reduce the number of change-overs, 

(3) reduce the number of film leaders and tail pieces required, and 

(4) reduce the amount of film wear. Briefly stated, the longer, or 
2000-foot, reel length retains and even exceeds the advantages ex- 
pected to be realized through the adoption of the 1700-foot length, 
without being subject to any of the latter's disadvantages. 

The opinion of this Committee relative to a change in the present 
reel length standard may be summarized as follows : 

(1) The Committee is not opposed to the present standard of 1000-foot reel 
length, if the exchanges and the theaters regard this length as a practical solution 
of the film reel problem. 

(2) The Committee emphatically is opposed to the proposed standard of a 
1700-foot maximum reel length, which, for the reasons previously cited, it does 
not regard as an appropriate solution to the problem. 

(5) If at any time in the future reel lengths exceeding 1000 feet are to be used, 
the Committee expresses its preference for, and endorsement of, the 2000-foot 
maximum and 1750-foot minimum lengths. 


The S. M. P. E. Standard Test Reel prepared by the Projection 
Practice Committee in cooperation with RCA- Victor Company is now 
available for general distribution. This reel, described in detail in the 
JOURNAL, 1 has been brought to the attention of the industry through 
the kind cooperation of Society members and industry trade publica- 
tions, for which aid the Committee expresses its appreciation. 

The response to the announcement of the availability of this reel 
has been gratifying, with reports from studios, theaters, and individ- 
uals indicating general satisfaction concerning its use. The Com- 


mittee invites comment and suggestions relative to the content and 
applicability of the reel, particularly with respect to any possible im- 


Work has been initiated and is progressing satisfactorily on a valu- 
able compendium of projection room practice and maintenance. 
Participation in this work is not limited to Committee members, and 
all contributions of data or suggestions will be appreciated. 


The Committee is engaged at present in an intensive study of new 
projection arc types, including the new a-c. arc. Papers embody- 
ing the results of this work will, it is hoped, be ready in the near future 
for presentation to the Society through the JOURNAL. 

H. RUBIN, Chairman 








1 Report of the Projection Practice Committee, /. Soc. Mot. Pict. Eng., XXI 
(Aug., 1933), No. 2, p. 89. 

Standard S. M. P. E. Visual and Sound Test Reels, /. Soc. Mot. Pict. Eng., 
XXII (March, 1934), No. 3, p. 173. 


MR. FINN: Some members may have gathered the impression that the Pro- 
jection Practice Committee has definitely committed itself to the a-c. arc or other 
arc types. That is not so; the Chairman of the Committee wishes to emphasize 
the fact that the Projection Practice Committee has not yet rendered any report 
on any new arcs. 

MR. FAULKNER: The rewind that is used in exchanges measures about 15 1 /* 
inches from the spindle to the table. The 15-inch reel allows about x /4 inch clear- 
ance. It would be possible to mount the rewind l /^ or l /z inch higher, on a block 
for the necessary clearance, but there would still be the gear ratio of 4 to 1 in the 
drive. The inspectress turns the handle of the rewind once, and the reel revolves 
four times. With 2000 feet of film on the take-up reel, or say, 1500 feet on the 
take-up reel and 500 on the feed reel, in order to detect the damaged places on the 
film she must exert so much pressure with her hand that the drag on the film 


becomes too great for her to turn the take-up with her right hand. Regardless 
of the size or the height of the rewind, the gear ratio is too great. Besides, the 
latter footage of the reel would be travelling so fast that she would not be able to 
detect the damage no matter how hard she tried. 

MR. QUINN: Film should not be examined at such a speed. That is, to ex- 
amine film properly (which is not the way it is done at the present time) the girl 
should stop at each splice, look at it, and see whether it has been made properly, 
and see whether the cement on the splice is holding. Allowing ten minutes for 
examining each 1000 feet of film, there should be no danger of her cutting her 
hand on any kind of reel that she might use, and the gear ratio of the rewind 
would not enter into consideration. 

MR. FAULKNER : The film is always inspected from the tail of the reel to the 
head, so that when the inspection is finished, the head of the reel is out ; therefore, 
the most important part of the film to be in good condition is the first 50 or 100 
feet. If 1500 to 2000 feet of film have been taken up on the right-hand reel, start- 
ing with the tail of the reel, and 150, 100, or 50 feet remain on the left-hand reel, 
even though the speed of inspection is reduced considerably, the latter footage of 
the reel will be travelling too fast. 

The human factor must be considered. If everything were done as it should 
be we probably shouldn't need these discussions. The average distributing 
company has from 250 to 300 inspectors who run the film through faster than 
they should. Again, in many instances much more work is demanded of them 
than they can properly do. If the exchange manager wants 50 reels a day, on 
the average, the inspectors must inspect that number; some exchanges 55, some 
60. Whether that is right or wrong is their problem. 

MR. QUINN: To provide new rewinds, changing the gear ratio to 3 to 1, or 1 to 
1, would cost only $3 at the most, if bought in quantity. The largest major 
company, Metro, for instance, has not more than 300 throughout the country. 
The rewinds would cost only $900; and they would not all have to be purchased 
at one time. 

MR. FAULKNER: Each inspector has two rewinds. That would be 600. 

MR. QUINN: Six hundred rewinds would not be very costly, in comparison 
with the good that would result from a 2000-ft. reel for the theater patrons; and 
after all, they are the ones who pay all our salaries and enable the industry to 
continue and prosper, yet they are the ones who are not considered. 

MR. FAULKNER: The rewind represents only a small item among the many 

MR. ROBIN: What consideration has been given to the load on the lamp, the 
carbon waste, the load on the rectifier or generating apparatus, in connection with 
the change in the reel length? 

MR. RUBIN: If the industry demands a change, if a change is necessary from 
the point of view of economy, as has been brought out, the Projection Practice 
Committee feels that the lamps, the rectifiers, the generators, and carbons will 
accommodate the 2000- as well as the 1000-ft. length. At the present time I do 
not know of any lamp that will not burn twenty minutes. I do not know of any 
generator or any rectifier that will not withstand twenty minutes of burning. 
Regarding the carbon waste, if such a standard is adopted, I am sure that the 
carbon companies will make their carbons conform to that standard. 


MR. GRIFFIN: All this controversy about the length of reels and the change in 
dimension of the reel came about through the fact that it is common practice 
throughout the country to double reels today. If that is true and doubling reels 
is common practice, the question raised by Mr. Robin can not enter into it. The 
equipment now is running for twenty or twenty-two minutes, and is standing up 
under the load. 

MR. QUINN: More than 90 per cent of the projectionists double the reels and 
run satisfactory shows. The National Carbon Company recently lengthened the 
13.6-mm. carbon from 20 to 22 inches, because of the fact that with double reels 
there was a waste of 2 inches. By adding 2 inches more it became possible to 
project an additional reel with each carbon. 

MR. CRABTREE: What is the real objection to the 1000-ft. reel? Is it that the 
projectionists waste too much film in splicing? 

MR. EDWARDS: Many reels are 400 feet or shorter. Now, while 400 feet of 
film is running it is obviously difficult, sometimes impossible, to retrim, rethread, 
and rewind; and those things have to be done right away. There are many 
theaters throughout the country in which there is only one projectionist, and he 
is kept extremely busy trying to keep the show running on 1000-ft. reels. The 
trouble is that there are rarely 1000 feet of film on a reel ; we can put on the show 
all right with 700 feet ; but with less there is difficulty. 

Another reason for doubling up is to provide a smoother performance. When 
you consider the length of time that the change-over dots appear on the screen, 
that is, the start-machine and the change-over dots, you only have to wink an 
eye, and they will be missed. There is always the possibility of a blank, which 
we always want to avoid. The 2000-ft. reel would reduce by 50 per cent the 
possibility of such errors. 

MR. CRABTREE: I am not objecting to the 2000 feet. Sometimes the pro- 
jectionist receives 1000-ft. reels and splices them up into 2000-ft. reels. What is 
the difference between his doing that and the exchange's supplying him with 
2000-ft. reels? Is it merely the time involved in splicing the reels and breaking 
them down when he returns them to the exchange? Does he spoil a lot of film 
in the process or does he not have sufficient time between receiving the film and 
putting it on the screen to do it properly? 

MR. EDWARDS: The objection to doubling is that every time that a cut is 
made, four frames are lost from every two reels two frames per reel that is, if 
the projectionist is careful. Unfortunately, the human factor enters, and some 
are not as careful as others. We are trying to evolve a scheme by which the man 
who has to project a film after 30 or 40 showings in different theaters will not lack 
the closing scenes entirely. 

MR. CRABTREE: How long is it before the film goes back to the exchange? 
Does it go through several theaters before it is returned to the exchange? 

MR. EDWARDS: Sometimes; sometimes there is a change every day. 

MR. CRABTREE: Why doesn't the exchange repair the damage? 

MR. EDWARDS: I can not answer that question. 

MR. HOLLANDER: It must be remembered that the several systems, such as 
the RCA and the Western Electric, operate at different speeds, pick-ups, starting 
points, and change-overs. To join part one to part two, and to cut two frames 
off, as Mr. Edwards said, will cause the loss of the change-over dots, which are 


only 15 frames from the end of the reel; and it takes only 7 theaters to cut off the 
change-over mark. The eighth projectionist who gets the print then makes his 
own change-over marks. He has a three-ft. pick-up; so he makes a 13-ft. mark 
instead of leaving the 10-ft. mark. If the film contained 2000 ft., and had a 
standard leader and a standard change-over, cutting the film would not be neces- 

MR. CRABTREE: But why do the exchanges allow the film to be clipped by 
seven persons before it is returned to them? 

MR. HOLLANDER: They can't prevent it. 

MR. FAULKNER: The change-over does not, of course, bother the projection- 
ists in the first-run houses. But by the time the film arrives at houses 30 to 60 
days later, the change-over cues are of little value. Each projectionist removes 
the head and tail pieces of the particular reels he wishes to double. 

Now, if a print goes out in seven reels, depending upon the program, one man 
might double a newsreel and reel No. 1. At the next theater the projectionist 
might double reel No. 1 and reel No. 2. Due to machine pick-ups, each projec- 
tionist makes his own change-over cues, and it often happens that in doubling 
reels he might wish to put a reel and a half on one reel and a reel and a half on 
the next. Then the cut occurs at the middle of the reel, and in order to change 
over, the projectionist would have to place his cue marks at that point. 

It is not a matter of what the exchange can do about it, nor is it the fault of the 
projectionist. Any condition under which the exchange handles its units in one 
length and the theaters in another is quite absurd. 

MR. RUBIN: Doubling reels was practiced in the days of the silent picture, 
when the standard length was 2000 feet. At the introduction of sound we had 
disks. The industry had to standardize on 1000-ft. reels as the disks were made 
to conform to only 1000 feet of film. Then came sound-on-film. The projec- 
tionist returned to his old length of 2000 feet, despite the fact that the standard 
for the change-over marks was 1000 feet. 

It is my understanding the producers or distributors feel that if the projectionist 
wants 2000 feet on the reel, and is determined to double, why not send out 2000-ft. 
reels, for the reasons stated by Mr. Edwards. However, the producers recom- 
mend rather a 1700-ft. reel, which means that on the average a theater will find it 
to be anywhere from 1200 feet to 1700 feet, and in most cases it will be 1200 feet. 

To cite an example: Bottoms Up, a recent picture, contained 7400 feet on 10 
reels you would not get 1700 feet, but much less. The reels are now averaging 
700 feet or less. If the standard is made 1700 feet the purpose of the new standard 
will be defeated because the projectionist is supplied with equipment that will still 
permit doubling. For that reason the Committee recommended that if a change 
is to be made, it should be to 2000 feet, with a minimum of 1750, and thus pre- 
vent doubling with his present equipment. 


The problem of recommending a design for a metal reel for mount- 
ing and maintaining release prints that might be standardized for 
use in the exchanges and would be acceptable to the projectionist 
for use in projection machines has been before the Committee. With 
the cooperation of representatives of the exchange operation depart- 
ments of the various distributing companies and members of the 
Projection Practice Committee, the matter of footage capacity of 
the reel was given the greatest consideration. 

The 1000-ft. reel, for which the present laboratory and exchange 
machines are geared, was compared with reels having greater footage 
capacity, as follows : 


(1) Saving of Film because of Fewer Change-Overs. The average 
print of seven reels mounted in 1000-ft. units necessitates "head and 
end" titles 8 feet long on each end of the reel, plus a change-over 
footage of 16 feet on each end, totalling 24 feet of film. To mount 
the print on reels having a capacity of 2000 feet, three such change- 
overs and 48 feet of film would be eliminated, or 144 feet of film to 
each print of seven reels. Based on an estimate of 500 releases by 
all companies annually, and 175 prints of each release, the footage 
saved would amount to 12,600,000 feet; or, on an estimated labora- 
tory charge of P/2 cents per foot, $189,000. 

(2) Prevention of Doubling of Reels by the Projectionist. A print 
edited and cut to 2000-ft. lengths would prevent the projectionist 
from doubling the reels, as the capacity of the projector magazines 
is not sufficient to accommodate the footage of two reels of film of 
that length. 

(3) Cost of Shipments. The possible reduction of the cost of ship- 
ping was discussed, but it was generally agreed that the total weight 
of a shipment comprising four 2000-ft. reels of film would be as great 
as one of seven reels of 1000-ft. capacity. 

* Presented at the Spring, 1934, Meeting at Atlantic City, N. J. 


Any capacity between 1000 feet and 2000 feet was unanimously 
disapproved, as a reel of 1500 feet or 1700 feet, as has been suggested 
at various times, would still offer the opportunity of doubling, as the 
projector magazine has a capacity of 3500 feet. 


(1) Short subjects of 1000 feet or less would either have to be 
mounted on large reels, or two sizes of reels and shipping cases be 
maintained for use in the exchange. It is estimated that more than 
50 per cent of the number of shipments made by all exchanges are 
individual shipments of 1000-ft. lengths or under. 

(2) Cost of Adapting Exchanges and Equipment for 2000- Ft. 

(a) Vault Racks. Few film vault racks could accommoda te shipping 
cases of the size required to store a 2000-ft. reel without expensive 

(b) Cost of Reels. Present reels of 1000-ft. capacity cost in the 
neighborhood of 12 cents each. A satisfactory 2000-ft. reel could 
probably be furnished, in quantity production, for 50 cents each. 
The 1000-ft. reels now in use would be discarded when final dis- 
position of the film mounted on them was made. 

(c) Shipping Cases. No noticeable difference in the cost of ship- 
ping cases is in favor or against 2000-ft. capacity. The saving in 
units about compensates for the cost of the additional strengthening 
required because of less compactness. The 1000-ft. shipping cases, 
discarded when the film they now hold is junked, would be a total 

(d) Rewinds. Present rewinds for 1000-ft. reels would be inade- 
quate for use with 2000-ft. reels as they are not sufficiently high for 
the larger reel. The gear ratio of 4 to 1 is too great to make it an 
easily operated unit for the average inspectress. 

(e) Film Bands. New film bands would have to be provided for 
the 2000-ft. reels. The only cost involved would be in discarding 
any stock of 1000-ft. bands on hand. 

(f) Vault Containers. Those companies using I.C.C. shipping 
cases for vault containers would not be affected; but where labora- 
tory tins are used for vault containers for 1000-ft. reels, the adoption 
of the 2000-ft. reels would require purchasing new or larger containers. 

(3) Inspection of Films. The use of 2000-ft. reels would present 
greater hazards, as regards injuring the inspectresses because of the 


greater weight. The outer rim of the metal reel would be travelling 
at a much greater speed when rewinding the latter footage of the reel 
and would, therefore, be harder to stop. 

Due to the higher rate of film travel, when inspecting the latter 
footage of the 2000-ft. reel, and remembering that the damaged parts 
are located by touch, it would be necessary that closer supervision 
be given to inspection to reduce the possibility of damaged portions 
of the film passing through the inspectors' fingers unnoticed. 

(4) Film Damage. A large proportion of film damage occurs in 
the projector. When such damage begins, usually the portion of 
the reel that follows is likewise damaged. The projectionist is not 
always aware that the damage is being done, and often when he is 
aware of it he can not afford to interrupt the show and, consequently, 
will not stop his machine. Where 2000-ft. reels are used, the damage 
done to the film throughout the remaining footage may be increased 
considerably if the damage begins somewhere in the first 1000 feet. 

Regardless of the size or capacity of the reel, the projectionist 
usually has his own personal 12 or 15 reels bought at a price far in 
excess of what the exchange can afford to pay for them. Because 
of the projectionist's confidence in the condition of his own reels, and 
his lack of confidence in the exchange's reels, the usual practice is to 
use the projection room reels regardless of the type or condition of 
the exchange reel. Therefore, the size or type or condition of the 
exchange reel has meaning only for the exchanges (inspection and 
shipping), except when lack of time makes it necessary for the pro- 
jectionist to run his first show "from the can." 

This practice in projection restricts the question of metal reels 
solely to exchange operation, but the matter of footage is of impor- 
tance to both projectionists and exchanges. The industry operates 
with the idea of delivering to its customers, through the medium of 
the screen, the ultimate in beautiful photography and continuity 
of sound and action. Therefore, the matter of reel footage is of more 
immediate concern to all of us than is the question of metal reels. 

From the view-point of what is best for the exchange, the latter 
would prefer the present 1000-ft. reels. But the projectionist has 
his rights in the matter; and after all, he is the medium through 
which the industry's efforts are delivered. For instance, one of the 
most popular releases of the past season was delivered to the pro- 
jection room mounted on 14 reels 14 thread-ups and 13 change - 
overs. The total footage of the subject was such that the pro- 


jectionist could mount it on four large reels. Due to the doubling 
of reels in the theaters during the early runs of the subject and their 
consequent mutilation, when they finally reached the one- and two-day 
theaters a series of change-over cues was scattered throughout, and so 
much footage lost at the original beginnings and ends of the reels 
that the screen result was anything but favorable. This, then, 
makes film footage of reels an item for exchange practice consideration. 

One of two things should be done. Either the projectionist should 
use the present 1000-ft. lengths in projection and not double the 
reels, or the film should be served to the theaters in lengths that will 
provide the best screen results, not only for the first-run theaters 
but for subsequent-run theaters, as well. 

To change over the studio editing and cutting departments so as 
to enable them to supply film to the exchanges in 2000-ft. lengths 
would : 

(1) Benefit the industry by saving an estimated footage of 12,600,- 
000 feet annually. 

(2) Eliminate the doubling of reels in projection, thus avoiding 
the consequent damage and loss of footage. 

To offset these benefits: 

(1) A different size of reel and shipping case would be required to 
handle subjects of 1000 feet or under. 

(2) Alterations in the vault racks would be required to accommo- 
date the larger storage can or shipping case. 

(3) The cost of reels for mounting new prints would be increased. 

(4) The loss of reels and shipping cases discarded when the 
1000-ft. units are retired would be entailed. 

(5) New vault containers would have to be purchased. 

(6) New rewinds would have to be purchased. 

(7) The danger of personal injury to the inspectresses would be 

() The efficiency of the inspectresses would be reduced. 

At this time, it is the belief of the Committee that the 1000-ft. 
reel now used in exchange practice is best suited for present exchange 
routines, as any metal reels furnished by the exchanges are rarely 
used by the projectionists. 

The economics of standardizing on the length of 2000 feet, or twice 
the footage of the present average reel, should be studied in all its 
phases. The efforts of the exchanges to maintain their prints prop- 
erly and the efforts of the projectionists to screen their pictures 


properly are apparently in opposition, each undoing what the other 
has done. If a cure for this condition exists it lies with the 2000-ft. 
reel, and not with the 1700-ft. reel with an average film footage on 
it of 1400 or 1500 feet, or any size reel that would permit doubling. 

T. FAULKNER, Chairman 




(This report was discussed jointly with that of the Projection Practice 
Committee, at the Spring, 1934, Meeting. The reader is therefore 
referred to p. 382 of this issue of the JOURNAL.) 


The regular monthly meeting, held in the Salle Moderne of the Hotel Pennsyl- 
vania, New York, on May 23rd, was attended by approximately 140 members 
and guests. Prior to the meeting, about 30 of the members met at an informal 
dinner in the Cafe of the hotel. It is planned next season to make this dinner 
preceding the meeting, to which all members and friends of the Section are 
cordially invited, a regular feature of the monthly meetings of the Section. 

The principal speaker of the evening was Mr. J. A. Norling, who presented a 
paper on "Methods of Process and Trick Cinematography," illustrated by a reel 
portraying many of the various fades, turn-overs, wipe-outs, etc., used in anima- 
tion and trick work. 

Following Mr. Norling's presentation, Mr. H. R. Kossman presented a reel 
of trick shots made in France and printed on a new Debrie automatic trick 

Further illustrations of animation and trick photography were illustrated in 
Brave Tin Soldier and Jack and the Beanstalk, colored cartoons supplied through 
the courtesy of Liberty Productions, Inc., and Popeye the Sailor, by Paramount 
Pictures Dist. Corp. 

The Section expresses its appreciation to Messrs. Griffin, Heidegger, and Knapp, 
of the International Projector Corp., for installing the projection equipment; 
and to the Raven Screen Corp. for presenting to the Society the screen used at 
the meeting. 


The regular monthly meeting of the Section was held on May 17th in Eckhart 
Hall, University of Chicago. Professor H. B. Lemon presented an interesting 
paper on "The Use of Sound Motion Pictures in Educational Work." Two new 
reels of sound pictures selected from the Physics Lectures of the University were 
presented: Sound Waves and Their Sources and Fundamentals of Acoustics. 
The meeting was well attended, and an interesting discussion followed. 


The regular monthly meeting of the Section was held on May 22nd, at the Don 
Lee Building, Hollywood, the subject being "Radio Television of Motion Pic- 

Starting at 8:00 P.M., the members were conducted on a tour of inspection, in 
groups, through the Don Lee studios and television transmitting stations W6XS 
and W6XAO, in operation. 

Convening again at 8:30 P.M., Mr. Harry R. Lubcke, director of television of 
the Don Lee Broadcasting System, described the equipment and commented on 
the relation of television to the motion picture art. The meeting terminated 
with a lively open-forum discussion of the subject by the members. 




On or about June 1st, the customary ballots will be mailed to the Honorary, 
Fellow, and Active members for nominations for officers of the Society for 1935. 
The Officers and Governors whose terms expire December 31, 1934, are as follows: 

*President: Alfred N. Goldsmith 
*Executive Vice-President : Harold C. Silent 

Editorial Vice-President: John I. Crabtree 

Convention Vice-President: William C. Kunzmann 
*Secretary: John H. Kurlander 
Treasurer: Timothy E. Shea 

Governor: Herford T. Cowling 

Governor: Ralph E. Farnham 
(Asterisks indicate one-year terms ; the remainder two-year terms.) 

At the next meeting of the Board of Governors, July 16th, the nominations 
returned by the members will be used in constructing the voting ballot to be mailed 
to the voting members about September 19th. The ballots will be counted at 
the Fall Convention, to be held in the Hotel Pennsylvania at New York, October 
29th-November 1st, and the results will be announced. The newly elected 
officers will assume their duties January 1, 1935. 



(and JENKINS, J. E.) 

(and SMITH, M. A.) 


(and JOY, D. B., 
and DOWNES, A. C.) 

(and SCHMIDT, R.) 



(and RAY, R. H.) 

(and JOY, D. B.) 

(and JOY, D. B., 

and BOWDITCH, F. T.) 



(and SANDVIK, O. t 

and HALL, V. C.) 
HALL, V. C. 

(and SANDVIK, O., 
and GRIMWOOD, W. K.) 


The Control Frequency Principle 

Some Practical Applications of Acoustics 
in Theaters 

Sixteen-Mm. Sound-on-Film 

High-Fidelity Lateral-Cut Disk Records 

The Economics of Projector Lamps for 
Advertising Purposes 

A New White Flame Carbon for Photo- 
graphic Light 

Two New Photographic Recording In- 
struments, the Sensitograph and the 

Equipment for Recording and Reproduc- 
ing Sound with Photofilm 

Sound Film Printing II 

Color for Industrial and Business Films 

Direct-Current High-Intensity Arcs with 
Non-Rotating Positive Carbons 

A New White Flame Carbon for Photo- 
graphic Light 

The Rotambulator A New Motion 
Picture Camera Stand 

A Study of Television Image Character- 

Transmission and Reproduction of Speech 
and Music in Auditory Perspective 

Sprocket Dimensions for 35-Mm. Visual 
and Sound Projection Equipment 

A New 35-Mm. Portable Sound Projector 

Further Investigation of Ground Noise 
in Photographic Sound Records 

Further Investigation of Ground Noise 
in Photographic Sound Records 

Issue Page 
Mar. 193 

Feb. 148 
Feb. 139 
Mar. 179 

Feb. 127 
Jan. 58 

May 279 

Mar. 157 
Apr. 215 
Feb. 98 
Feb. 144 

Jan. 42 
Jan. 58 

Mar. 200 
May 290 
May 314 

Jan. 20 
Jan. 70 

Feb. 83 

Feb. 83 





(and MITCHELL, R. F.) 

(and ADAIR, S. E.) 
JOY, D. B. 

(and DOWNES, A. C.) 
JOY, D. B. 

(and BOWDITCH, F. T., 

and DOWNES, A. C.) 


(and HOWELL, A. S.) 

RAY, R. H. 

(and CRESS, H. W.) 
ROTH, G. E. 


(and HALL, V. C., 
and GRIMWOOD, W. K.) 

(and BRANDES, H.) 


(and BAKER, G. W.) 
WOLF, S. K. 

Wide-Range Recording 

Recent Improvements in the Bell 

Howell Fully Automatic Printer 
The Control Frequency Principle 


Issue Page 
Apr. 253 

Feb. 115 
Mar. 193 

Direct-Current High-Intensity Arcs with 

Non-Rotating Positive Carbons Jan. 42 

A New White Flame Carbon for Photo- 
graphic Light Jan. 58 

The Use of the Talking Picture as an 
Additional Educational Tool at the 
University of Chicago Jan. 62 

Recent Improvements in the Bell & 

Howell Fully Automatic Printer Feb. 115 

A New Development in Carbon Arc 

Lighting Jan. 

An Automatic Change-Over Device Mar. 

Color for Industrial and Business Films Feb. 

The "Selenophon" Sound Recording and 

Reproducing System Apr. 260 

Further Investigation of Ground Noise 

in Photographic Sound Records Feb. 83 

Two New Photographic Recording In- 
struments, the Sensitograph and the 
Gammagraph May 279 

Some Practical Applications of Acoustics 

in Theaters Feb. 148 

Acoustical Requirements for Wide-Range 

Reproduction of Sound Apr. 242 





Some Practical Applications of Acoustics in Theaters, G. W. Baker and M. A. 

Smith, No. 2 (Feb.), p. 148. 
Acoustical Requirements for Wide-Range Reproduction of Sound, S. K. Wolf, 

No. 4 (April), p. 242. 
Advertising Equipment 

The Economics of Projector Lamps for Advertising Purposes, E. W. Beggs, 

No. 2 (Feb.), p. 127. 
The Control Frequency Principle, J. E. Jenkins and S. E. Adair, No. 3 (March), 

p. 193. 

A New Development in Carbon Arc Lighting, P. Mole, No. 1 (Jan.), p. 51. 
Recent Improvements in the Bell & Howell Fully Automatic Printer, A. S. 

Howell and R. F. Mitchell, No. 2 (Feb.), p. 115. 

Standard S. M. P. E. Visual and Sound Test Reels, No. 3 (March), p. 173. 
An Automatic Change-Over Device, A. Pritchard, No. 3 (March), p. 186. 
The Rotambulator A New Motion Picture Camera Stand, J. A. Dubray, 

No. 3 (March), p. 200. 

A New 35-Mjn. Portable Sound Projector, H, Griffin, No. 1 (Jan.), p. 70. 
Two New Photographic Recording Instruments, the Sensitograph and the 

Gammagraph, H. Brandes and R. Schmidt, No. 5 (May), p. 279. 
Arcs, Projection 

Direct-Current High-Intensity Arcs with Non-Rotating Positive Carbons, 

D. B. Joy and A. C. Downes, No. 1 (Jan.), p. 42. 
Auditory Perspective 

Transmission and Reproduction of Speech and Music in Auditory Perspec- 
tive, H. Fletcher, No. 5 (May), p. 314. 

An Automatic Change-Over Device, A. Pritchard, No. 3 (March), p. 186. 
Color Cinematography 

Color for Industrial and Business Films, R. H. Ray and H. W. Cress, No. 2 
(Feb.), p. 144. 

Committee Reports 

Report of the Committee on Laboratory and Exchange Practice, No. 1 (Jan.), 

p. 3; No. 5 (May), p. 332. 

Report of the Sub-Committee on Exchange Practice, No. 6 (June), p. 386. 
Report of the Projection Practice Committee, No. 1 (Jan.), p. 11; No. 2 (Feb.), 
p. 153; No. 3 (March), p. 212; No. 5 (May), p. 332; No. 6 (June), p. 379. 
Report of the Historical and Museum Committee, No. 1 (Jan.), p. 13. 


396 INDEX [j. s. M. P. E. 

Report of the Committee on Standards and Nomenclature, No. 1 (Jan.), 

p. 17; No. 1 (Jan.), p. 79; No. 5 (May), p. 333. 
Progress in the Motion Picture Industry: Report of the Progress Committee, 

No. 6 (June), p. 341. 
Membership Committee, No. 5 (May), p. 333. 

Disk Recording 

High-Fidelity Lateral-Cut Disk Records, F. C. Barton, No. 3 (March), p. 179. 

Educational Cinematography 

The Use of the Talking Picture as an Additional Educational Tool at the Uni- 
versity of Chicago, H. B. Lemon, No. 1 (Jan.), p. 62. 

Exchange Practice 

Report of the Committee on Laboratory and Exchange Practice, No. 1 (Jan.), 
p. 3; No. 6 (June), p. 386. 


Two New Photographic Recording Instruments, the Sensitograph and the 
Gammagraph, H. Brandes and R. Schmidt, No. 5 (May), p. 279. 


The Use of the Talking Picture as an Additional Educational Tool at the Uni- 
versity of Chicago, H. B. Lemon, No. 1 (Jan.), p. 62. 

The Economics of Projector Lamps for Advertising Purposes, E. W. Beggs, 
No. 2 (Feb.), p. 127. 

Equipment for Recording and Reproducing Sound with Photofilm, A. F. 
Chorine, No. 3 (March), p. 157; No. 4 (April), p. 215. 

The Control Frequency Principle, J. E. Jenkins and S. E. Adair, No. 3 (March), 
p. 193. 

Transmission and Reproduction of Speech and Music in Auditory Perspec- 
tive, H. Fletcher, No. 5 (May), p. 314. 

Ground Noise 

Further Investigation of Ground Noise in Photographic Sound Records, 
O. Sandvik, V. C. Hall, and W. K. Grimwood, No. 2 (Feb.), p. 83. 


Report of the Historical and Museum Committee, No. 1 (Jan.), p. 13. 

Illumination in Photography 

A New Development in Carbon Arc Lighting, P. Mole, No. 1 (Jan.), p. 51. 
A New White Flame Carbon for Photographic Light, D. B. Joy, F. T. Bow- 
ditch, and A. C. Downes, No. 1 (Jan.), p. 58. 

Incandescent Lamps 

The Economics of Projector Lamps for Advertising Purposes, E. W. Beggs, 
No. 2 (Feb.), p. 127. 


Author Index, Vol. XXII, Jan. to June, 1934; No. 6 (June), p. 393. 
Classified Index, Vol. XXII, Jan. to June, 1934; No, 6 (June), p. 395. 

June, 1934] INDEX 397 

Industrial Cinematography 

Color for Industrial and Business Films, R. H. Ray and H. W. Cress, No. 2 

(Feb.), p. 144. 
Instruments, Measuring 

Standard S. M. P. E. Visual and Sound Test Reels, No. 3 (March), p. 173. 
Two New Photographic Recording Instruments, the Sensitograph and the 
Gammagraph, H. Brandes and R. Schmidt, No. 5 (May), p. 279. 

Laboratory Practice 

Report of the Committee on Laboratory and Exchange Practice, No. 1 (Jan.), 

p. 3. 

The Economics of Projector Lamps for Advertising Purposes, E. W. Beggs, 

No. 2 (Feb.), p. 127. 

A New Development in Carbon Arc Lighting, P. Mole, No. 1 (Jan.), p. 51. 
A New White Flame Carbon for Photographic Light, D. B. Joy, F. T. Bow- 
ditch, and A. C. Downes, No. 1 (Jan.), p. 58. 
Local Sections 

Atlantic Coast Section, No. 1 (Jan.), p. 79; No. 2 (Feb.), p. 153; No. 3 (March), 

p. 212; No. 5 (May), p. 331; No. 6 (June), p. 391. 
Mid-West Section, No. 1 (Jan.), p. 79; No. 2 (Feb.), p. 153; No. 4 (April), 

p. 271; No. 5 (May), p. 331; No. 6 (June), p. 391. 

Pacific Coast Section, No. 1 (Jan.), p. 78; No. 4 (April), p. 271; No. 6 (June), 
p. 391. 


Rectification of S. M. P. E. Membership, No. 3 (March), p. 209. 

Report of the Historical and Museum Committee, No. 1 (Jan.), p. 13. 


Peter A. Snell, No. 5 (May), p. 333. 
Open Forum 
Should Studio Recording Equipment Compensate for Theater Reproducing 

Characteristics? No. 3 (March), p. 183. 
Organization of the Society 

Functional and Administrative Organization of the S. M. P. E., No. 4 (April), 
p. 270. 

Portable Equipment 

A New 35-Mm. Portable Sound Projector, H. Griffin, No. 1 (Jan.), p. 70. 
Sixteen-Millimeter Sound-on-Film, J. O. Baker, No. 2 (Feb.), p. 139. 


Sound Film Printing II, J. Crabtree, No. 2 (Feb.), p. 98. 
Recent Improvements in the Bell & Howell Fully Automatic Printer, A. S. 
Howell and R. F. Mitchell, No. 2 (Feb.), p. 115. 

398 INDEX [j. s. M. P. E. 

Processing, Control of 

Two New Photographic Recording Instruments, the Sensitograph and the 

Gammagraph, H. Brandes and R. Schmidt, No. 5 (May), p. 279. 

Progress in the Motion Picture Industry: Report of the Progress Committee, 

No. 6 (June), p. 341. 
Projection, General Information 

Sprocket Dimensions for 35-Mm. Visual and Sound Projection Equipment, 

H. Griffin, No. 1 (Jan.), p. 20. 
Direct- Current High-Intensity Arcs with Non-Rotating Positive Carbons, 

D. B. Joy and A. C. Downes, No. 1 (Jan.), p. 42. 

A New 35-Mm. Portable Sound Projector, H. Griffin, No. 1 (Jan.), p. 70. 
Report of the Projection Practice Committee, No. 1 (Jan.), p. 11; No. 6 

(June), p. 379. 

Standard S. M. P. E. Visual and Sound Test Reels, No. 3 (March), p. 173. 
An Automatic Change-Over Device, A. Pritchard, No. 3 (March), p. 186. 
Projection Practice 

Report of the Projection Practice Committee, No. 1 (Jan.), p. 11; No. 6 

(June), p. 379. 
Standard S. M. P. E. Visual and Sound Test Jleels, No. 3 (March), p. 173. 


The "Selenophon" Sound Recording and Reproducing System, G. E. Roth, 

No. 4 (April), p. 260. 

Two New Photographic Recording Instruments, the Sensitograph and the 

Gammagraph, H. Brandes and R. Schmidt, No. 5 (May), p. 279. 
Sixteen-Millimeter Equipment 

Sixteen-Mm. Sound-on-Film, J. O. Baker, No. 2 (Feb.), p. 139. 
Sound Recording 

Further Investigation of Ground Noise in Photographic Sound Records, 

O. Sandvik, V. C. Hall, and W. K. Grimwood, No. 2 (Feb.), p. 83. 
Equipment for Recording and Reproducing Sound with Photofilm, A. F. Chor- 
ine, No. 3 (March), p. 157; No. 4 (April), p. 215. 

High-Fidelity Lateral-Cut Disk Records, F. C. Barton, No. 3 (March), p. 179. 
Open Forum: Should Studio Recording Equipment Compensate for Theater 

Reproducing Characteristics? No. 3 (March), p. 183. 
Wide-Range Recording, F. L. Hopper, No. 4 (April), p. 253. 
The "Selenophon" Sound Recording and Reproducing System, G. E. Roth, 

No. 4 (April), p. 260. 
Two New Photographic Recording Instruments, the Sensitograph and the 

Gammagraph, H. Brandes and R. Schmidt, No. 5 (May), p. 279. 
Sound Reproduction 

Equipment for Recording and Reproducing Sound with Photofilm, A. F. 

Chorine, No. 3 (March), p. 157; No. 4 (April), p. 215. 

Open Forum : Should Studio Recording Equipment Compensate for Theater 
Reproducing Characteristics? No. 3 (March), p. 183, 

June, 1934] INDEX 

Acoustical Requirements for Wide-Range Reproduction of Sound, S. K. Wolf, 

No. 4 (April), p. 242. 
The "Selenophon" Sound Recording and Reproducing System, G. E. Roth, No. 

4 (April), p. 260. 
Transmission and Reproduction of Speech and Music in Auditory Perspective, 

H. Fletcher, No. 5 (May), p. 314. 
Sprocket Dimensions for 35-Mm. Visual and Sound Projection Equipment. 

H. Griffin, No. 1 (Jan.), p. 20. 

Report of the Committee on Standards and Nomenclature, No. 1 (Jan.), p. 17. 
Sprocket Dimensions for 35-Mm. Visual and Sound Projection Equipment, 

H. Griffin, No. 1 (Jan.), p. 20. 

Standard S. M. P. E. Visual and Sound Test Reels, No. 3 (March), p. 173. 
Studio Equipment 

A New Development in Carbon Arc Lighting, P. Mole, No. 1 (Jan.), p. 51. 
A New White Flame Carbon for Photographic Light, D. B. Joy, F. T. Bow- 
ditch, and A. C. Downes, No. 1 (Jan.), p. 58. 
The Rotambulator A New Motion Picture Camera Stand, J. A. Dubray, 

No. 3 (March), p. 200. 

A Study of Television Image Characteristics, E. W. Engstrom, No. 5 (May), 

p. 290. 
Test Reel 

Standard S. M. P. E. Visual and Sound Test Reels, No. 3 (March), p. 173. 
Theater Characteristics 

Some Practical Applications of Acoustics in Theaters, G. W. Baker and M. A. 

Smith, No. 2 (Feb.), p. 148. 
Open Forum: Should Studio Recording Equipment Compensate for Theater 

Reproducing Characteristics? No. 3 (March), p. 183. 
Acoustical Requirements for Wide-Range Reproduction of Sound, S. K. Wolf, 

No. 4 (April), p. 242. 
Transmission of Sound 

Transmission and Reproduction of Speech and Music in Auditory Perspective, 
H. Fletcher. No. 5 (May), p. 314. 



Prepared under the Supervision 



Two reels, each approximately 500 feet long, of specially pre- 
pared film, designed to be used as a precision instrument in 
theaters, review rooms, exchanges, laboratories, and the like 
for testing the performance of projectors. The visual section 
includes special targets with the aid of which travel-ghost, 
lens aberration, definition, and film weave may be detected 
and corrected. The sound section includes recordings of 
various kinds of music and voice, in addition to constant 
frequency, constant amplitude recordings which may be used 
for testing the quality of reproduction, the frequency range 
of the reproducer, the presence of flutter and 60-cycle or 96- 
cycle modulation, and the adjustment of the sound track. 
Reels sold complete only (no short sections). 


(Shipped to any point in the United States) 
Address the